Skip to content

What can the M4 Max 128 GB run?

What fits on the M4 Max 128 GB
Text models: 119Visual models: 9Strongest first
Memory
128 GB unified
Bandwidth
546 GB/s
Maker
Apple

Open the calculator on this machine

Text models that run on it

Text models that run on the M4 Max 128 GB, best first
ModelBest fitContextSpeedStrongest at
Qwen3.8 Flash-Next3-bit256K71 tok/sChat, Coding, Agents
MiniMax M31.6-bit64K50 tok/sDocs
Qwen3.8 27B8-bit256K13 tok/sChat, Coding, Agents, Docs
Qwen3.5 397B A17B1.6-bit256K57 tok/sAgents, Docs
MiniMax M2.72-bit40K56 tok/sCoding, Agents
Muse Glimmer 30B8-bit128K13 tok/sDocs
Mistral Medium 3.5 128B5-bit16K4.4 tok/sChat, Coding, Agents, Docs
Qwen3.5 122B A10B5-bit256K41 tok/sChat, Coding, Agents, Docs
Nemotron 3 Super 120B A12B5-bit256K30 tok/sChat, Coding, Agents, Docs
Nemotron 3.5 Bolt 74B A9B8-bit256K28 tok/sChat, Coding, Agents, Docs
Qwen3.6 35B A3B8-bit256K74 tok/sChat, Coding, Agents, Docs
Gemma 4 31B8-bit256K12 tok/sChat, Coding, Agents, Docs
Gemma 4 26B A4B8-bit256K71 tok/sChat, Coding, Agents, Docs
GLM-4.7 Flash8-bit198K70 tok/sChat, Coding, Agents
Mistral Small 4 119B5-bit256K61 tok/sChat, Docs
gpt-oss 120BMXFP4128K72 tok/sChat, Coding
Qwen3 Coder Next 80B8-bit256K68 tok/sCoding, Agents
Nemotron 3.5 Lightning 30B A3B8-bit256K66 tok/sAgents, Docs
Devstral Small 2 24BFP8384K16 tok/sCoding
LFM2.5 24B A2B8-bit128K93 tok/sAgents
Mistral Small 3.2 24B8-bit128K16 tok/sChat, Coding, Agents
Llama 3.3 70B8-bit64K5.4 tok/sChat, Coding, Agents, Docs
Olmo 3.1 32B8-bit64K11 tok/sChat, Coding, Agents, Docs
Qwen3 32B8-bit40K11 tok/sNot rated
Qwen3.6 27B8-bit256K13 tok/sText and images
Qwen3 Coder 30B A3B Instruct8-bit256K69 tok/sCoding
Qwen3.5 27B8-bit256K13 tok/sText and images
Qwen3.5 35B A3B8-bit256K74 tok/sText and images
NVIDIA Nemotron 3 Nano 30B A3B BF168-bit256K66 tok/sNot rated
GLM 4.5 Air6-bit32K27 tok/sNot rated
Inkling Small2-bit4K61 tok/sText and images
Qwen3 30B A3B8-bit40K69 tok/sNot rated
MiniMax M2.52-bit40K52 tok/sNot rated
Qwen3 30B A3B Instruct 25078-bit256K69 tok/sNot rated
Laguna XS.28-bit256K74 tok/sNot rated
OTel 2.0 LLM 31B IT8-bit256K12 tok/sNot rated
Laguna M.12-bit40K32 tok/sNot rated
Qwen2.5 32B Instruct8-bit32K11 tok/sNot rated
MiniCPM V 4.7 35B A3B8-bit256K74 tok/sText and images
Laguna XS 2.18-bit256K74 tok/sNot rated
Diffusiongemma 26B A4B IT8-bit256K69 tok/sText and images
Qwen2.5 VL 32B Instruct8-bit125K11 tok/sText and images
Qwen2.5 Coder 32B Instruct8-bit32K11 tok/sCoding
Laguna S 2.16-bit16K42 tok/sNot rated
Qwen3 VL 30B A3B Instruct8-bit256K69 tok/sText and images
DeepSeek R1 Distill Qwen 32B8-bit128K11 tok/sNot rated
Gemma 3 27B IT8-bit128K14 tok/sText and images
Qwen3 VL 235B A22B Instruct2-bit44K30 tok/sText and images
Qwen3 235B A22B2-bit40K30 tok/sNot rated
Ornith 1.5 35B A3B8-bit256K74 tok/sNot rated
Ornith 1.5 122B5-bit256K34 tok/sNot rated
Ornith 1.5 397B1.6-bit256K47 tok/sNot rated
Qwen3 VL 32B Instruct8-bit224K11 tok/sText and images
Step 3.5 Flash2-bit52K67 tok/sNot rated
Qwen2.5 72B Instruct8-bit32K5.2 tok/sNot rated
Qwen3 Next 80B A3B Instruct8-bit256K68 tok/sNot rated
Llama 4 Scout 17B 16E Instruct6-bit36K23 tok/sText and images
MiniMax M22-bit40K52 tok/sNot rated
Qwen2.5 VL 72B Instruct8-bit56K5.2 tok/sText and images
Qwen1.5 7B8-bit32K48 tok/sNot rated

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 119 are shown.

Image and video models

Image and video models that run on the M4 Max 128 GB
ModelBest fitPer outputStrongest at
MiniMax H316-bit62 minVideo
LTX-2.516-bit6.5 minVideo
Qwen-Image16-bit37 minImages
FLUX.2 [dev]16-bit17 minImages
Wan2.2 I2V A14B16-bit5.5 hVideo
Wan2.1 I2V 14B16-bit5.5 hVideo
Wan 2.1 T2V-14B16-bit6.9 hVideo
Qwen Image 251216-bit37 minImages
Wan 2.2 A14B16-bit5.5 hVideo

Only just too big

These need up to a quarter more memory than the M4 Max 128 GB has. Spilling into system RAM or a lower precision can get them running, slowly.

Questions

What is the best LLM for the M4 Max 128 GB?

Of the open models we track, the strongest that run well on the M4 Max 128 GB are Qwen3.8 Flash-Next, MiniMax M3, Qwen3.8 27B. 119 text models fit it in all.

Calculated from published specs, not benchmarked. All machines