Skip to content

Mistral Small 3.1 24B Instruct 2503

Mistral AI
Code Multilingual Tool Calls Vision

Mistral Small 3.1 24B Instruct 2503 is a 24-billion-parameter dense transformer from Mistral AI, optimized for instruction following, code generation, vision understanding, and multilingual conversation. It succeeds the earlier Mistral Small 2501 with an expanded 128K context window and a native Pixtral vision encoder. The model supports tool calling and 24 languages including English, French, Arabic, and Hindi. With flash attention and efficient quantization, it fits on a single consumer GPU at Q4 for self-hosted inference.

Hardware Configuration

Optional — for precise deployment recommendations
Quantization Quality Size Fit
Q8_0 High 23.33 GB
Q8_K_XL High 27 GB
Q6_K High 18.02 GB
Q6_K_XL High 19.36 GB
Q5_K_M Medium 15.61 GB
Q5_K_S Medium 15.18 GB
Q5_K_XL Medium 15.61 GB
Q4_K_M Medium 13.35 GB
Q4_K_S Medium 12.62 GB
Q4_K_XL Medium 13.51 GB
Q4_0 Medium 12.57 GB
Q4_1 Medium 13.85 GB
Q3_K_M Low 10.69 GB
Q3_K_S Low 9.69 GB
Q3_K_XL Low 11.04 GB
Q2_K Low 8.28 GB
Q2_K_L Low 8.43 GB
Q2_K_XL Low 8.65 GB
Last updated: March 7, 2026