Can the Apple M4 Max run Llama 3.3 70B?
Yes, with 64 GB or more. Llama 3.3 70B needs about 45 GB and the Apple M4 Max with 64 GB of unified memory leaves about 48 GB for models. Expect about 7 tok/s, slow, but it works. The 36 and 48 GB versions don't have room.
ollama run llama3.3:70bFirst time? Install Ollama, then paste that into a terminal.
Each Apple M4 Max version
- 36 GB (about 24 GB usable)Too big
- 48 GB (about 36 GB usable)Too big
- 64 GB (about 48 GB usable)Slow ~7 tok/s
- 128 GB (about 96 GB usable)Slow ~7 tok/s
Speed comes from memory bandwidth: the card reads all 43 GB of active weights for every token. On the Apple M4 Max, gpt-oss 120B runs at about 66 tok/s if you can trade some quality for speed. Cards that run Llama 3.3 70B faster:
Speed is roughly memory bandwidth divided by the weights read per token, calibrated on our own rack. Sizes are Ollama's default 4-bit downloads plus about 1.2 GB for context. Real results are usually within about a third. Measured it yourself? Send us the number.