What can the Mac mini M4 Pro 64 GB run?
| Text models: 181 | Visual models: 20 | Strongest first |
- Memory
- 64 GB unified
- Bandwidth
- 273 GB/s
- Maker
- Apple
Text models that run on it
| Model | Best fit | Context | Speed | Strongest at |
|---|---|---|---|---|
| Qwen3.8 27B | 8-bit | 256K | 7.6 tok/s | Chat, Coding, Agents, Docs |
| Muse Glimmer 30B | 8-bit | 128K | 7.2 tok/s | Docs |
| Qwen3.5 122B A10B | 2-bit | 40K | 41 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3 Super 120B A12B | 2-bit | 208K | 30 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3.5 Bolt 74B A9B | 4-bit | 256K | 27 tok/s | Chat, Coding, Agents, Docs |
| Qwen3.6 35B A3B | 8-bit | 256K | 50 tok/s | Chat, Coding, Agents, Docs |
| Gemma 4 31B | 8-bit | 256K | 6.5 tok/s | Chat, Coding, Agents, Docs |
| Gemma 4 26B A4B | 8-bit | 256K | 45 tok/s | Chat, Coding, Agents, Docs |
| GLM-4.7 Flash | 8-bit | 198K | 48 tok/s | Chat, Coding, Agents |
| Gemma 4 12B | 8-bit | 256K | 17 tok/s | Chat, Coding, Docs |
| Mistral Small 4 119B | 2-bit | 144K | 63 tok/s | Chat, Docs |
| gpt-oss 120B | 2-bit | 112K | 74 tok/s | Chat, Coding |
| Qwen3 Coder Next 80B | 3-bit | 256K | 73 tok/s | Coding, Agents |
| Nemotron 3.5 Lightning 30B A3B | 8-bit | 256K | 46 tok/s | Agents, Docs |
| Devstral Small 2 24B | FP8 | 128K | 8.7 tok/s | Coding |
| LFM2.5 24B A2B | 8-bit | 128K | 66 tok/s | Agents |
| Mistral Small 3.2 24B | 8-bit | 128K | 8.8 tok/s | Chat, Coding, Agents |
| Ministral 3 14B | 8-bit | 192K | 15 tok/s | Coding, Agents, Docs |
| gpt-oss 20B | MXFP4 | 128K | 67 tok/s | Agents |
| Ministral 3 8B | 8-bit | 256K | 24 tok/s | Docs |
| Llama 3.3 70B | 4-bit | 16K | 5.1 tok/s | Chat, Coding, Agents, Docs |
| Olmo 3.1 32B | 8-bit | 64K | 6.4 tok/s | Chat, Coding, Agents, Docs |
| Llama 3.1 8B | 8-bit | 128K | 26 tok/s | Chat, Coding, Agents, Docs |
| Qwen3 14B | 8-bit | 40K | 14 tok/s | Not rated |
| Qwen3 VL 8B Instruct | 8-bit | 256K | 24 tok/s | Text and images |
| Qwen3 8B | 8-bit | 40K | 26 tok/s | Not rated |
| Qwen3 32B | 8-bit | 40K | 6.4 tok/s | Not rated |
| Qwen3.6 27B | 8-bit | 256K | 7.6 tok/s | Text and images |
| Qwen2.5 Coder 14B Instruct | 8-bit | 32K | 14 tok/s | Coding |
| Qwen3 Coder 30B A3B Instruct | 8-bit | 152K | 48 tok/s | Coding |
| Qwen3.5 27B | 8-bit | 256K | 7.6 tok/s | Text and images |
| Qwen3.5 35B A3B | 8-bit | 256K | 50 tok/s | Text and images |
| NVIDIA Nemotron 3 Nano 30B A3B BF16 | 8-bit | 256K | 46 tok/s | Not rated |
| GLM 4.5 Air | 2-bit | 32K | 32 tok/s | Not rated |
| Qwen3 30B A3B | 8-bit | 40K | 48 tok/s | Not rated |
| Qwen3 30B A3B Instruct 2507 | 8-bit | 152K | 48 tok/s | Not rated |
| Laguna XS.2 | 8-bit | 256K | 50 tok/s | Not rated |
| OTel 2.0 LLM 31B IT | 8-bit | 256K | 6.5 tok/s | Not rated |
| Qwen2.5 32B Instruct | 8-bit | 32K | 6.4 tok/s | Not rated |
| MiniCPM V 4.7 35B A3B | 8-bit | 256K | 50 tok/s | Text and images |
| Qwen2.5 14B Instruct | 8-bit | 32K | 14 tok/s | Not rated |
| Laguna XS 2.1 | 8-bit | 256K | 50 tok/s | Not rated |
| Diffusiongemma 26B A4B IT | 8-bit | 256K | 43 tok/s | Text and images |
| Qwen2.5 VL 32B Instruct | 8-bit | 44K | 6.2 tok/s | Text and images |
| Qwen2.5 Coder 32B Instruct | 8-bit | 32K | 6.4 tok/s | Coding |
| Laguna S 2.1 | 2-bit | 80K | 48 tok/s | Not rated |
| Qwen3 VL 30B A3B Instruct | 8-bit | 144K | 48 tok/s | Text and images |
| EXAONE 3.5 7.8B Instruct | 8-bit | 32K | 27 tok/s | Not rated |
| DeepSeek R1 0528 Qwen3 8B | 8-bit | 128K | 26 tok/s | Not rated |
| Gemma 3 12B IT | 8-bit | 128K | 18 tok/s | Text and images |
| Phi 3.5 Mini Instruct | 8-bit | 104K | 49 tok/s | Not rated |
| DeepSeek R1 Distill Qwen 32B | 8-bit | 48K | 6.4 tok/s | Not rated |
| Gemma 3 27B IT | 8-bit | 128K | 7.8 tok/s | Text and images |
| Ornith 1.5 35B A3B | 8-bit | 256K | 50 tok/s | Not rated |
| Ornith 1.5 122B | 2-bit | 40K | 35 tok/s | Not rated |
| GPT Neox 20B | 8-bit | 2K | 10.0 tok/s | Not rated |
| Apertus 8B Instruct 2509 | 8-bit | 64K | 26 tok/s | Not rated |
| Phi 4 | 8-bit | 16K | 14 tok/s | Not rated |
| Phi 3 Mini 128K Instruct | 8-bit | 104K | 49 tok/s | Not rated |
| Qwen3 VL 32B Instruct | 8-bit | 44K | 6.3 tok/s | Text and images |
Best fit is the highest quality, up to 8-bit, that leaves the use case’s context. Ranked by our use-case scores, then by Hugging Face downloads. Speeds are rough, plus or minus 25%. The top 60 of 181 are shown.
Image and video models
| Model | Best fit | Per output | Strongest at |
|---|---|---|---|
| MiniMax H3 | 4-bit | 2.4 h | Video |
| LTX-2.5 | 8-bit | 13 min | Video |
| Z-Image Turbo | 16-bit | 1.9 min | Images |
| FLUX.1 [dev] | 16-bit | 3.9 min | Images |
| Qwen-Image | 8-bit | 75 min | Images |
| FLUX.2 [dev] | 5-bit | 38 min | Images |
| Qwen-Image 2.1 | 16-bit | 32 min | Images |
| FLUX.3 Klein | 16-bit | 2.9 min | Images |
| Wan2.2 I2V A14B | 16-bit | 11.0 h | Video |
| FLUX.2 [klein] 9B | 16-bit | 55 s | Images |
| Wan2.1 I2V 14B | 16-bit | 11.0 h | Video |
| Wan 2.1 T2V-14B | 16-bit | 13.8 h | Video |
| Krea 2 Turbo | 16-bit | 3.9 min | Images |
| Stable Diffusion 3.5 Large | 16-bit | 17 min | Images |
| Qwen Image 2512 | 8-bit | 75 min | Images |
| MOSS-Image | 16-bit | 21 min | Images |
| Wan 2.2 TI2V-5B | 16-bit | 3.0 h | Video |
| MOSS Image Base | 16-bit | 21 min | Images |
| Wan 2.2 A14B | 16-bit | 11.0 h | Video |
| HunyuanVideo 1.5 | 16-bit | 13.5 h | Video |
Only just too big
These need up to a quarter more memory than the Mac mini M4 Pro 64 GB has. Spilling into system RAM or a lower precision can get them running, slowly.
Questions
What is the best LLM for the Mac mini M4 Pro 64 GB?
Of the open models we track, the strongest that run well on the Mac mini M4 Pro 64 GB are Qwen3.8 27B, Muse Glimmer 30B, Qwen3.5 122B A10B. 181 text models fit it in all.
Calculated from published specs, not benchmarked. All machines