Apple M2 Max
96 GB decides what fits. 400 GB/s decides how fast it runs once it does.
Computed for 96 GB, the largest configuration. The Apple M2 Max is also sold with 32 or 64 GB, and what fits changes with it.
Run it with 32 GB →Run it with 32 GB →Run it with 64 GB →Run it with 64 GB →
Open in the calculator →Best models for 96 GB, on every card that size →
Find one: Amazon ↗ · eBay (new and used) ↗Store links may pay us a commission. They never decide which card is suggested — the memory arithmetic does.
- Largest popular model that fitsLing-3.0-flash127B params · Q4_K_M · needs 78.2 GB~72 tok/s Faster than you readOpen in the calculator →
- Most downloaded that fitsQwen3.8-27B28B params · Q4_K_M · needs 18.6 GB~22 tok/s About reading paceOpen in the calculator →
At 8,192 tokens of context with the whole model in device memory. Speeds are estimates from memory bandwidth, not benchmarks run on this card.
245 of 320 fit entirely
245 of the 320 models the engine can size fit entirely in device memory at Q4_K_M where it is published, otherwise the nearest published format, and 8,192 tokens. Unified memory is one pool, so there is no second memory tier to spill into; the remaining 75 do not run at this context.
Featured models · Q4_K_M at 8,192 tokens, including offload
| Model | Needs | Verdict | Decode | Calculator |
|---|---|---|---|---|
| Qwen3.5-2B2.3B parameters | 2.32 GB | fits | ~111 tok/s | Open → |
| NVIDIA-Nemotron-3-Nano-4B-BF164.0B parameters | 3.46 GB | fits | ~78 tok/s | Open → |
| Qwen3-VL-8B-Instruct8.8B parameters | 7.04 GB | fits | ~47 tok/s | Open → |
| Qwen3.5-9B9.7B parameters | 7.05 GB | fits | ~54 tok/s | Open → |
| gemma-4-12B-it12B parameters | 9.54 GB | fits | ~36 tok/s | Open → |
| gpt-oss-20b21B parameters | 13.8 GB | fits | ~81 tok/s | Open → |
| Qwen3.8-27B28B parameters | 18.6 GB | fits | ~22 tok/sAbout reading pace | Open → |
| Qwen3.6-27B28B parameters | 18.6 GB | fits | ~22 tok/sAbout reading pace | Open → |
| Kimi-Linear-48B-A3B-Instruct49B parameters | 30.4 GB | fits | ~94 tok/s | Open → |
| Qwen3.8-Flash-Next180B parameters | 111.6 GB | short by 15.6 GB | does not run | Open → |
| DeepSeek-V4-Flash-Vision-Exp305B parameters | 188.1 GB | short by 92.1 GB | does not run | Open → |
| GLM-5.3-Flash321B parameters | 198.3 GB | short by 102.3 GB | does not run | Open → |
These examples are selected from prominent labs using the catalogue’s latest Hugging Face 30-day downloads and repository-creation freshness signal, with newer releases guaranteed a place. Offloaded rows assume 32 GB of system RAM, and a speed is only shown for a row that runs. Fitting in memory is not the same as loading: whether the runtime and version you have supports each architecture and format on this machine has not been tested here. Each “Open” link carries the same model, format, context and RAM into the calculator.
Fits entirely in Apple M2 Max (M2 Max 12-core CPU / 38-core GPU, 96 GB) memory — 245 of 320 sized models
The most demanding model that fits is Mixtral-8x22B-Instruct-v0.1 at Q4_K_M: 87.7 GB of the 96 GB, leaving 8.29 GB spare.
Fully resident at 8,192 tokens (or the model’s own maximum, where that is shorter), offload off, at 400 GB/s. Each row is a run of the engine for this configuration; the rows start with current, prominent releases and “fits” is memory, not a tested runtime. Older or less prominent models remain available through this search and “Show all”.
Showing 40 of 245 models that fit.
| Model | Format | Needs | Spare | Decode | Downloads / 30d | Calculator |
|---|---|---|---|---|---|---|
| Qwen3.8-27BQwen · 28B params | Q4_K_M | 18.6 GB | 77.4 GB | ~22 tok/sAbout reading pace | 6.9M | Open → |
| gemma-4-26B-A4B-itGoogle · 26B params | Q4_K_M | 16.8 GB | 79.2 GB | ~78 tok/s | 13M | Open → |
| gemma-4-31B-itGoogle · 31B params | Q4_K_M | 22.2 GB | 73.8 GB | ~16 tok/sAbout reading pace | 9.9M | Open → |
| Qwen3.5-9BQwen · 9.7B params | Q4_K_M | 7.05 GB | 89.0 GB | ~54 tok/s | 9M | Open → |
| Qwen3.5-4BQwen · 4.7B params | Q4_K_M | 3.98 GB | 92.0 GB | ~78 tok/s | 7.8M | Open → |
| Qwen3.6-35B-A3BQwen · 36B params | Q4_K_M | 23.1 GB | 72.9 GB | ~94 tok/s | 3.3M | Open → |
| gemma-4-12B-itGoogle · 12B params | Q4_K_M | 9.54 GB | 86.5 GB | ~36 tok/s | 1.9M | Open → |
| Qwen3.6-27BQwen · 28B params | Q4_K_M | 18.6 GB | 77.4 GB | ~22 tok/sAbout reading pace | 2.5M | Open → |
| Qwen3.5-2BQwen · 2.3B params | Q4_K_M | 2.32 GB | 93.7 GB | ~111 tok/s | 4.9M | Open → |
| gemma-4-E4B-itGoogle · 8.0B params | Q4_K_M | 5.86 GB | 90.1 GB | ~73 tok/s | 4.4M | Open → |
| NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16NVIDIA · 32B params | Q4_K_M | 20.3 GB | 75.7 GB | ~17 tok/sAbout reading pace | 530K | Open → |
| NVIDIA-Nemotron-3-Nano-4B-BF16NVIDIA · 4.0B params | Q4_K_M | 3.46 GB | 92.5 GB | ~78 tok/s | 3.4M | Open → |
| gemma-4-E2B-itGoogle · 5.1B params | Q4_K_M | 4.02 GB | 92.0 GB | ~114 tok/s | 3M | Open → |
| Qwen3.5-0.8BQwen · 873M params | Q4_K_M | 1.46 GB | 94.5 GB | ~139 tok/s | 2.6M | Open → |
| Qwen3-VL-8B-InstructQwen · 8.8B params | Q4_K_M | 7.04 GB | 89.0 GB | ~47 tok/s | 14.6M | Open → |
| Qwen3.5-27BQwen · 28B params | Q4_K_M | 18.6 GB | 77.4 GB | ~22 tok/sAbout reading pace | 1.9M | Open → |
| North-Micro-Vision-InstructCohereLabs · 2.5B params | Q4_K_M | 2.99 GB | 93.0 GB | ~94 tok/s | 180.7K | Open → |
| Qwen3.5-35B-A3BQwen · 36B params | Q4_K_M | 23.1 GB | 72.9 GB | ~94 tok/s | 1.6M | Open → |
| NVIDIA-Nemotron-3-Super-120B-A12B-BF16NVIDIA · 124B params | Q4_K_M | 76.9 GB | 19.1 GB | ~4.8 tok/sSlow | 1.2M | Open → |
| granite-4.2-8bIBM · 8.8B params | Q4_K_M | 7.49 GB | 88.5 GB | ~44 tok/s | 128.4K | Open → |
| GLM-4.7-Flashzai-org · 31B params | Q4_K_M | 20.4 GB | 75.6 GB | ~17 tok/sAbout reading pace | 1.8M | Open → |
| granite-4.1-3bIBM · 3.4B params | Q4_K_M | 3.72 GB | 92.3 GB | ~76 tok/s | 521.8K | Open → |
| LFM2.5-2.6BLiquidAI · 2.7B params | Q4_K_M | 2.61 GB | 93.4 GB | ~97 tok/s | 108.1K | Open → |
| Qwen3-VL-4B-InstructQwen · 4.4B params | Q4_K_M | 4.72 GB | 91.3 GB | ~64 tok/s | 3.5M | Open → |
| granite-4.1-30bIBM · 29B params | Q4_K_M | 20.4 GB | 75.6 GB | ~17 tok/sAbout reading pace | 301.5K | Open → |
| Qwen3.5-122B-A10BQwen · 125B params | Q4_K_M | 77.9 GB | 18.1 GB | ~52 tok/s | 512.6K | Open → |
| Qwen3-VL-2B-InstructQwen · 2.1B params | Q4_K_M | 3.05 GB | 93.0 GB | ~90 tok/s | 2.8M | Open → |
| granite-4.2-3bIBM · 3.7B params | Q4_K_M | 3.72 GB | 92.3 GB | ~76 tok/s | 50.6K | Open → |
| LFM2.5-230MLiquidAI · 230M params | Q4_K_M | 1.05 GB | 94.9 GB | ~158 tok/s | 88.6K | Open → |
| Qwen3-0.6BQwen · 752M params | Q4_K_M | 2.22 GB | 93.8 GB | ~108 tok/s | 29.7M | Open → |
| Qwen3-Coder-NextQwen · 80B params | Q4_K_M | 50.0 GB | 46.0 GB | ~76 tok/s | 596.3K | Open → |
| gpt-oss-20bOpenAI · 21B params | Q4_K_M | 13.8 GB | 82.2 GB | ~81 tok/s | 6.6M | Open → |
| granite-4.2-30bIBM · 29B params | Q4_K_M | 20.7 GB | 75.3 GB | ~17 tok/sAbout reading pace | 35.8K | Open → |
| granite-4.1-8bIBM · 8.8B params | Q4_K_M | 7.49 GB | 88.5 GB | ~44 tok/s | 179.3K | Open → |
| NVIDIA-Nemotron-3-Nano-30B-A3B-BF16NVIDIA · 32B params | Q4_K_M | 20.3 GB | 75.7 GB | ~17 tok/sAbout reading pace | 875.2K | Open → |
| gpt-oss-120bOpenAI · 117B params | Q4_K_M | 71.9 GB | 24.1 GB | ~66 tok/s | 4.5M | Open → |
| LFM2.5-VL-3BLiquidAI · 3.1B params | Q4_K_M | 2.85 GB | 93.1 GB | ~97 tok/s | 26.7K | Open → |
| MiMo-V2.6-Distill-Qwen-9BNEWXiaomiMiMo · 9.4B params | Q4_K_M | 6.90 GB | 89.1 GB | ~54 tok/s | 14.6K | Open → |
| Qwen3-4B-Instruct-2507Qwen · 4.0B params | Q4_K_M | 4.72 GB | 91.3 GB | ~64 tok/s | 3.7M | Open → |
| Ling-3.0-tinyinclusionAI · 7.9B params | Q4_K_M | 5.85 GB | 90.1 GB | ~125 tok/s | 17.7K | Open → |
Sized at the format most people actually download, not at FP16. The catalogue ordered by downloads →
Find the Apple M2 Max: Amazon ↗ · eBay (new and used) ↗Store links may pay us a commission. They never decide which card is suggested — the memory arithmetic does.
The specification behind every figure
What the manufacturer publishes for this device, and the pages it was read from.
Manufacturer specification
| Memory scope | unified-system |
|---|---|
| Capacity | 96 GB |
| Published options | 32 GB, 64 GB, 96 GB |
| Bandwidth | 400 GB/s |
| Memory type | unified memory |
| Bus width | Not published |
| FP32 peak | Not published |
| Dense matrix peak | Not published without sparsity |
| Power | Not published |
Source ledger
Caveats
- The 30-core GPU M2 Max's 400GB/s is printed on the 14-inch MacBook Pro (2023) page; the 38-core GPU's 400GB/s on the 16-inch MacBook Pro (2023) page ("Apple M2 Max chip 12-core CPU … 38-core GPU 16-core Neural Engine 400GB/s memory bandwidth"). 96GB is sold only with the 38-core GPU.
- Unified memory is shared with the operating system and applications; their use is not subtracted.
- Apple publishes no peak throughput figure for this chip, so no compute roof is priced for it and time to first token is withheld.
Official configurations
| Configuration | GPU cores | Memory | Bandwidth |
|---|---|---|---|
| M2 Max 12-core CPU / 30-core GPU | 30 | 32 / 64 GB | 400 GB/s |
| M2 Max 12-core CPU / 38-core GPU | 38 | 32 / 64 / 96 GB | 400 GB/s |
Published capacity is a hardware ceiling, not guaranteed free runtime memory. The calculator shows the runtime reserve separately rather than folding it into a single number.
Run the diagnostic on the Apple M2 Max →