Any repository / resolved against 327 catalogued models

Paste a Hugging Face link

Most downloads are fine-tunes and GGUF files of a model, not the model itself. Paste the repository you are about to download and see which catalogued model answers for it, and why.

How a repository is answered

Its publisher’s own base_model tag is followed, through a GGUF repository to the fine-tune it quantizes if need be, until it names a model the catalogue holds. A repository’s name is never read as evidence of what it is built on.
Its config.json is compared with the base’s pinned one on every field that decides memory: layers and their pattern, widths, head counts, experts, latent and recurrent dimensions, vocabulary and any vision or audio tower.
If they match, the base model’s memory and speed answers are the repository’s, and the files it publishes are listed with their own sizes. If they differ, or nothing was published to compare, the page says which.

Derivatives already compared

68 repositories, checked 2026-10-03

Answered by	Repositories with the same geometry
Qwen3-VL-2B-Instruct	typhoon-ai/typhoon-ocr1.5-2b fine-tune
Qwen3.8-27B	autotrust/JEV-27B-VL adapter ukisai/Swift-Qwen3.8-27b fine-tune Youssofal/Ternary-Bonsai-2-27B-MTPLX-Optimized-Speed merge Cloudflare/clef fine-tune tsinghua-sigs-robot-lab/VeriLoop-E2 fine-tune bytkim/Qwen3.8-27B-pi fine-tune BAAI/AREX-2 fine-tune OmniJev/OneJev-27B fine-tune
Qwen3-0.6B	PrimeIntellect/Qwen3-0.6B fine-tune superwhisper/s1-mini fine-tune
gpt-oss-20b	FastFlowLM/GPT-OSS-20B-NPU2 quantization
Mistral-7B-v0.1	mistralai/Mistral-7B-Instruct-v0.1 fine-tune HuggingFaceH4/zephyr-7b-beta fine-tune NousResearch/Hermes-2-Pro-Mistral-7B fine-tune, context or vocabulary differs teknium/OpenHermes-2.5-Mistral-7B fine-tune, context or vocabulary differs HuggingFaceH4/zephyr-7b-alpha fine-tune Intel/neural-chat-7b-v3-1 fine-tune
gemma-4-26B-A4B-it	google/gemma-4-26B-A4B-it-qat-q4_0-unquantized fine-tune hiwaifu-research/WaifuGemma4-26b-a4b-v1 fine-tune
gemma-4-31B-it	google/gemma-4-31B-it-qat-q4_0-unquantized fine-tune
Qwen3.5-0.8B	ATH-MaaS/OvisOCR2 fine-tune togethercomputer/Tev1-0.8B-experimental fine-tune internlm/Intern-Decision-0.8B fine-tune SeerRay-Lab/Xiaomi-OCR-0 fine-tune
Qwen2.5-VL-7B-Instruct	nvidia/Cosmos-Reason1-7B fine-tune reducto/RolmOCR fine-tune Hcompany/Holo1-7B fine-tune
Qwen3.5-27B	Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled fine-tune microsoft/Fara1.5-27B fine-tune
Qwen3.5-9B	empero-ai/Qwythos-9B-Claude-Mythos-5-1M fine-tune, context or vocabulary differs Cloudflare/clef-flash fine-tune OrionLLM/OxCoder-9B fine-tune emperorofrome/Gmcoder merge Tesslate/OmniCoder-9B fine-tune
Qwen3-VL-30B-A3B-Instruct	browser-use/bu-30b-a3b-preview fine-tune
Qwen3.5-35B-A3B	apodex/Apodex-1.1-mini fine-tune Hcompany/Holo3-35B-A3B fine-tune
Qwen3.5-4B	togethercomputer/Tev1-4B-experimental fine-tune nvidia/NV-Reason-CT fine-tune internlm/Intern-Decision-4B fine-tune kirp/jpt-4b fine-tune
Qwen3.6-35B-A3B	juspay/xor fine-tune Accio-Lab/occamy-1.0 fine-tune endless-frontier/BigBang-v1 fine-tune
MiMo-V2.6-Distill-Qwen-9B	Hikari07jp/MiMo-V2.6-Distill-Qwen-9B-Ablitrated fine-tune
DeepSeek-R1-Distill-Qwen-1.5B	agentica-org/DeepScaleR-1.5B-Preview fine-tune
Qwen3.8-Flash-Next	ukisai/Swift1.5-Qwen3.8-Flash-Next fine-tune badtheorylabs/Tinfield-1 fine-tune Blackfrost-AI/CYBER-FROST-3.8-BF16 fine-tune beamster/Qwen3.8-Flash-Next-Sushi-4bpw quantization, context or vocabulary differs beamster/Qwen3.8-Flash-Next-Sushi-2.6bpw quantization, context or vocabulary differs beamster/Qwen3.8-Flash-Next-Sushi-2bpw quantization, context or vocabulary differs
DeepSeek-V3.2-Exp	deepseek-ai/DeepSeek-V3.2-Speciale fine-tune
Qwen3-8B	nvidia/Nemotron-Orchestrator-8B fine-tune
llm-jp-4.1-32b-a3b-thinking	Hikari07jp/llm-jp-4.1-32b-a3b-thinking-Ablitrated quantization
granite-4.2-3b	webAI-Official/TwIL-LM3-Pro adapter
K2-Horizon-0.9B	IFM/K2-Type-0.9B fine-tune
Qwen3.6-27B	JetBrains/Qwen3.8-3.6-27B-blend merge
gemma-4-E2B-it	RinggAI/ringg-router-e2b fine-tune
Qwen3.5-2B	internlm/Intern-Decision-2B fine-tune
DeepSeek-R1-Distill-Qwen-14B	agentica-org/DeepCoder-14B-Preview fine-tune
Mistral-7B-Instruct-v0.2	segolilylabs/Lily-Cybersecurity-7B-v0.2 fine-tune
Qwen3-4B	Menlo/Jan-nano fine-tune
GLM-5.3-Flash	cantina-security/apex-flash-1 fine-tune
Qwen2.5-32B-Instruct	NovaSky-AI/Sky-T1-32B-Preview fine-tune
Qwen3.5-397B-A17B	prefeitura-rio/Rio-3.5-Open-397B fine-tune
Kolibri-1-BF16	Aleph-Alpha/Kolibri-1 quantization

These are the derivatives with enough downloads or attention to appear in the daily coverage audit, whose configuration it compared with the base’s and found to match. Any other repository can be pasted above and is compared on request.