Magistral Small 1.1Open Weight

Released

Magistral Small 1.1 is an AI model released by Mistral on Thursday, Jul 24 2025, 14 days after Devstral Medium. It is an open-weight model — the trained weights are available to download and run. It is a 24B parameterThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. model with a 40k token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover LiveCodeBench, AIME 2024, AIME 2025, and GPQA Diamond.

Benchmarks

Coding

LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better.
59.17%
#36 of 60

Reasoning & science

AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools
70.52%
#13 of 21
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools
62.03%
#27 of 29
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better.
65.78%
#94 of 125

Compare Magistral Small 1.1 with

Magistral Small 1.1

About

Magistral Small 1.1 was released on July 24, 2025 as an update to Mistral's downloadable 24B reasoning model. Mistral described broadly similar benchmark performance to version 1.0, with improvements to tone, Markdown and LaTeX formatting, shorter replies to straightforward prompts and fewer generation loops. The update focused on making the model's reasoning and answers easier to use.

Mistral published the weights under Apache 2.0. They supported a 128K-token context, but the company recommended a 40K working limit because performance could degrade beyond it. The recorded AIME, GPQA Diamond and LiveCodeBench v5 results belong to this July checkpoint. Vision input arrived with the later 1.2 update, so Small 1.1 remained a text reasoning model.

Frequently asked questions

Magistral Small 1.1 was released by Mistral on Thursday, Jul 24 2025.

All Mistral releases

45 tracked