作者： Jakub Rusinowski · 最后更新： 2026年4月5日
Yes, comfortably — you'll have ~60 GB of headroom running Devstral-2 123B at Q4_K_M (68 GB, ~25 tok/s (est.)) with room for up to 4K context.
在亚马逊查看价格 — Apple Mac Studio M4 Max
| VRAM | 128 GB unified memory |
| Memory Bandwidth | 546 GB/s |
| Quant | VRAM needed | Fits 128 GB? | Max context |
|---|---|---|---|
| F16 | 286.2 GB | ✗ No | — |
| Q8_0 | 170.8 GB | ✗ No | — |
| Q6_K | 141 GB | ✗ No | — |
| Q5_K_M | 127.3 GB | ✓ Yes | 4K |
| Q4_K_M | 114.4 GB | ✓ Yes | 4K |
| Q3_K_M | 92.6 GB | ✓ Yes | 4K |
| Q2_K | 80.6 GB | ✓ Yes | 8K |
VRAM needed assumes a 4K-token context with an f16 KV cache; “Max context” is the largest window that still fits in 128 GB of unified memory. Figures are estimates from parameter count, quantization and memory bandwidth — the analyzer lets you tune KV-cache quant and context.
| Devstral-2 123B | Q4_K_M · 68 GB · ~25 tok/s (est.) |
| Devstral-2 22B | Q4_K_M · 13 GB · ~42 tok/s (est.) |
At 2 hrs/day, buying (~$3,499) beats renting at $0.77/hr after about 6.3 years.
Affiliate links — we may earn a commission if you sign up, at no extra cost to you.
Cloud rates verified 2026-07 — estimates, and marketplace prices vary. Buying price is GPU MSRP only, not a full PC.
Yes, comfortably — you'll have ~60 GB of headroom running Devstral-2 123B at Q4_K_M (68 GB, ~25 tok/s (est.)) with room for up to 4K context.
Devstral-2 123B at Q4_K_M quantization (68 GB), estimated ~25 tokens/sec, up to 4K context.
← Can I Run It? | Devstral-2 Model Page | Apple M4 Max GPU Page | Check Your Hardware