Mistral AI

Devstral Small 2 24B

code

24B

Parameters

13,3 Go

Q4_K_M file

≥ 14,8 Go

Recommended VRAM

393 216

Max context

Not measured yet

No card has run the TokenBench protocol on this model yet. The current protocol covers a selection of models; benchmarking any catalog model from the app is coming soon.

Measure my card →

Which cards should it fit on?

Cards on the site whose video memory covers the recommended 14,8 Go (file + 4,000-token context). An estimate, not a measurement. Click a card for the detailed estimate.

Technical sheet

Published: 2025-11-28
Architecture: 40 layers · 8 KV heads · dimension 128
Attention: Standard: every layer keeps the whole context in memory.
Context cache (f16): ≈ 156 MB per 1,000 tokens · 5,0 Go for 32,768 tokens
GGUF file: bartowski/mistralai_Devstral-Small-2-24B-Instruct-2512-GGUF · mistralai_Devstral-Small-2-24B-Instruct-2512-Q4_K_M.gguf Download (13,3 Go)
Official repo: mistralai/Devstral-Small-2-24B-Instruct-2512

Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.