Devstral Small 24B — VRAM Requirements
Written by Jakub Rusinowski · Last updated August 15, 2026
How much GPU VRAM you need to run Devstral Devstral Small 24B by Mistral AI locally, a 24B-parameter model. Figures are quantized weights + KV cache + framework overhead, computed from the model's parameter count and published architecture — not a throughput model. See /en/methodology.
Devstral Small 24B needs about 17 GB VRAM at Q4_K_M.
VRAM by Quantization
| Quant | Bits/weight | Weights | Total VRAM |
|---|---|---|---|
| Q2_K | 2.63 | 7.9 GB | 10.0 GB |
| Q3_K_M | 3.41 | 10.2 GB | 12.4 GB |
| Q4_K_M | 4.83 | 14.5 GB | 16.6 GB |
| Q5_K_M | 5.67 | 17.0 GB | 19.2 GB |
| Q6_K | 6.56 | 19.7 GB | 21.8 GB |
| Q8_0 | 8.50 | 25.5 GB | 27.6 GB |
| F16 | 16.00 | 48.0 GB | 50.1 GB |
Switch quantization in the interactive calculator, or see the full Devstral model page.
Add this badge to your model card
Model creators: paste this into your Hugging Face model card README to link readers straight to this VRAM breakdown.
[](https://llmconfigurator.com/en/vram-calculator/devstral-small-24b?utm_source=badge&utm_medium=referral&utm_campaign=readme_badge&utm_content=devstral-small-24b)Estimates only — actual VRAM varies with context length, batch size, runtime and KV-cache settings.