Written by Jakub Rusinowski · Last updated August 15, 2026
Yes, but it is tight
Yes, but it is tight — Devstral-2 123B at Q8_0 needs about 134.4 GB of the 144 GB usable on Mac Studio (M2 Ultra, 192 GB), leaving only ~9.6 GB before the runtime starts swapping. Expect ~9.7 tok/s (estimated), with room for about 32,768 tokens of context.
Confidence: low · Recommended quantization: Q8_0 · Estimated speed: ~9.7 tok/s
| Usable memory for models | 144 GB |
| Memory bandwidth | 800 GB/s |
| Form factor | Desktop |
| Operating system | macOS |
| Memory upgradeable | No — soldered |
| Quant | Memory needed | Fits 144 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 249.7 GB | ✗ No | — | — | 246 GB |
| Q8_0 | 134.4 GB | ✓ Yes | 32K | ~9.7 tok/s | 130.7 GB |
| Q6_K | 104.5 GB | ✓ Yes | 64K | ~12.3 tok/s | 100.9 GB |
| Q5_K_M | 90.9 GB | ✓ Yes | 64K | ~14.1 tok/s | 87.2 GB |
| Q4_K_M | 77.9 GB | ✓ Yes | 64K | ~16.3 tok/s | 74.3 GB |
| Q3_K_M | 56.1 GB | ✓ Yes | 64K | ~22.1 tok/s | 52.4 GB |
| Q2_K | 44.1 GB | ✓ Yes | 64K | ~27.6 tok/s | 40.4 GB |
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| Devstral-2 123B | 77.9 GB | ✓ Fits | ~16.3 tok/s |
| Devstral Small 24B | 16.6 GB | ✓ Fits | ~26.6 tok/s |
| Devstral-2 22B | 15.7 GB | ✓ Fits | ~28.5 tok/s |
Ollama or LM Studio (Metal) — MLX for the fastest Apple-native throughput
Yes, but it is tight — Devstral-2 123B at Q8_0 needs about 134.4 GB of the 144 GB usable on Mac Studio (M2 Ultra, 192 GB), leaving only ~9.6 GB before the runtime starts swapping. Expect ~9.7 tok/s (estimated), with room for about 32,768 tokens of context.
Q8_0 — it needs about 134.4 GB of the 144 GB available, downloads as roughly 130.7 GB, and runs at an estimated 9.7 tokens/sec with up to 32K of context.
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Ollama or LM Studio (Metal) — MLX for the fastest Apple-native throughput