Can the Apple M4 Max (128GB) run Mixtral 8x7B?
Yes
QuantFP16Full quality
Memory95.1 GBof 96 GB
Speed~13tokens/s
Yes. The Apple M4 Max (128GB) can run Mixtral 8x7B fully in unified memory at FP16 (full quality), using about 95.1 GB of the 96 GB available, at roughly 13 tokens/second.
Try other context lengths in the calculator →Memory needed by quantization
At 8,192 tokens of context. macOS lets the GPU use about 96 GB of the 128 GB unified memory by default.
| Quant | Quality | Memory | Fits? | Speed |
|---|---|---|---|---|
| FP16 | Full quality | 95.1 GB | Fits | ~13 tok/s |
| Q8_0 | Near-lossless | 51.3 GB | Fits | ~24 tok/s |
| Q6_K | Excellent | 40 GB | Fits | ~31 tok/s |
| Q5_K_M | Very good | 34.9 GB | Fits | ~36 tok/s |
| Q4_K_M | Good, most popular | 30 GB | Fits | ~42 tok/s |
| Q3_K_M | Noticeably degraded | 24.5 GB | Fits | ~52 tok/s |
| Q2_K | Heavily degraded | 19.2 GB | Fits | ~68 tok/s |
How to run Mixtral 8x7B on the Apple M4 Max (128GB)
With Ollama installed, one command downloads and starts it:
ollama run mixtral:8x7bOllama's default download is Q4_K_M (about 30 GB here). You have room for FP16, which is full quality: pick that GGUF in LM Studio or from Hugging Face.
GPUs that run Mixtral 8x7B comfortably
Models that run well on the Apple M4 Max (128GB)
Run it in the cloud
Don't want to buy hardware? Run Mixtral 8x7B on a rented GPU today.
Rent a cloud GPU by the hour. No upfront cost, cancel any time.