Can the Apple M4 run Llama 3.3 70B?
No. Llama 3.3 70B needs about 45 GB and the Apple M4 has at most 32 GB of unified memory, of which about 21 GB is usable for models. The most capable model it runs well is Qwen3 30B-A3B, at about 21 tok/s.
Llama 3.3 70B needsAbout 45 GB
Apple M4 has16 or 24 or 32 GB unified memory
Bandwidth120 GB/s
Estimated speedDoesn't fit
By memory size
Each Apple M4 version
- 16 GB (about 11 GB usable)Too big
- 24 GB (about 16 GB usable)Too big
- 32 GB (about 21 GB usable)Too big
What runs Llama 3.3 70B
These cards have the memory for it, nearest to the Apple M4 first. The cheapest build on our list that runs it well is the Workstation + 2× RTX 3090, from $3,500 used, at about 12 tok/s.
How we estimate
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.
Keep going