Written by Jakub Rusinowski · Last updated August 15, 2026
Yes, comfortably — you'll have ~21.7 GB of headroom running Devstral-2 123B at Q4_K_M (74.26125 GB, ~52 tok/s (est.)) with room for up to 32K context.
or compare on Vast.ai from $0.77/hr (typical low · varies)
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
| VRAM | 96 GB |
| Memory Bandwidth | 1792 GB/s |
| Quant | VRAM needed | Fits 96 GB? | Max context |
|---|---|---|---|
| F16 | 248.2 GB | ✗ No | — |
| Q8_0 | 132.9 GB | ✗ No | — |
| Q6_K | 103.1 GB | ✗ No | — |
| Q5_K_M | 89.4 GB | ✓ Yes | 16K |
| Q4_K_M | 76.5 GB | ✓ Yes | 32K |
| Q3_K_M | 54.7 GB | ✓ Yes | 64K |
| Q2_K | 42.7 GB | ✓ Yes | 64K |
VRAM needed assumes a 4K-token context with an f16 KV cache; “Max context” is the largest window that still fits in 96 GB. 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 · 74.26125 GB · ~52 tok/s (est.) |
| Devstral Small 24B | Q4_K_M · 14.49 GB · ~82 tok/s (est.) |
| Devstral-2 22B | Q4_K_M · 13.2825 GB · ~87 tok/s (est.) |
At 2 hrs/day, buying (~$8,565) beats renting at $0.77/hr after about 15.4 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 ~21.7 GB of headroom running Devstral-2 123B at Q4_K_M (74.26125 GB, ~52 tok/s (est.)) with room for up to 32K context.
Devstral-2 123B at Q4_K_M quantization (74.26125 GB), estimated ~52 tokens/sec, up to 32K context.
← Can I Run It? | Devstral Model Page | NVIDIA RTX PRO 6000 Blackwell GPU Page | Check Your Hardware