Can I run Mixtral 8x7B (MoE) on Apple M2 (16GB)?
Not comfortably: Mixtral 8x7B (MoE) only runs on Apple M2 (16GB) with Q2_K by offloading part of the model to system RAM, which is much slower (often a few tokens/s). Smaller models or lower quantisations fit entirely in memory.
Memory by quantisation (8,192-token context)
| Quantisation | Memory | Verdict | Est. speed |
|---|---|---|---|
| F16Unquantized reference | 89.6 GiB | Does not fit | — |
| Q8_0Near-lossless | 48.8 GiB | Does not fit | — |
| Q6_KVery close to F16 | 38.3 GiB | Does not fit | — |
| Q5_K_MGreat quality/size | 33.5 GiB | Does not fit | — |
| Q4_K_MBest default for most | 29.0 GiB | Does not fit | — |
| IQ4_XSSmaller 4-bit (imatrix) | 25.7 GiB | Needs CPU offload | — |
| Q3_K_MNoticeable loss | 23.9 GiB | Needs CPU offload | — |
| Q2_KLast resort; big loss | 20.8 GiB | Needs CPU offload | — |
Apple M2 (16GB) offers about 16 GiB of memory (67% usable for a model) at 100 GB/s. Speed is a bandwidth-bound estimate of single-stream generation at 65% efficiency; real results vary by runtime, drivers and settings, so measure on your machine.
How this is calculated
Weights ≈ parameters × bits per weight ÷ 8. KV cache ≈ 2 × 32 layers × 8 KV heads × 128 head dim × 2 bytes per token. Decode speed ≈ efficiency × memory bandwidth ÷ bytes read per token (this is a mixture-of-experts model, so only the active parameters are read, but all of them must fit in memory). Learn the maths in VRAM math and memory bandwidth, or try the VRAM calculator.
Mixtral 8x7B (MoE) on other hardware
Similar models on Apple M2 (16GB)
Running slowly or out of memory? See CUDA out of memory and the quantisation lesson.