Magistral Medium 1.1

Released

Magistral Medium 1.1 is an AI model released by Mistral on Thursday, Jul 24 2025, the same day as Magistral Small 1.1. It has 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.
59.35%
#35 of 60

Reasoning & science

AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools
72.03%
#11 of 21
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools
60.99%
#28 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.
71.46%
#84 of 125

Compare Magistral Medium 1.1 with

Magistral Medium 1.1

About

Magistral Medium 1.1 was announced in Mistral's July 24, 2025 model update as the hosted counterpart to Magistral Small 1.1. It continued the Medium reasoning line introduced in June, giving API users an updated model for tasks that benefited from longer deliberation.

The release had a documented 40K-token context and remained a commercial service without publicly downloadable weights. Mistral evaluated it separately from both the original Medium and the smaller July checkpoint. The version suffix therefore matters when comparing reasoning results, especially across the later September update that also added image understanding.

Frequently asked questions

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

All Mistral releases

45 tracked