Benchmark Almanac

Magistral Small 1.0

Magistral Small 1.0.

Mistral AI Released June 10th, 2025 Compare with others
14
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Mistral AI Jun 2025

Magistral Small 1.0

Open weights

129.9 #275 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
  • MMMU-Pro 51.8% #77 /90
  • GPQA Diamond 56.1% #145 /229
  • LiveCodeBench v5 55.8% #34 /47

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

BenchmarkScoreRankSource
Humanity's Last Exam 6.40% #151 of 154 Lab model card

Math and science

BenchmarkScoreRankSource
MATH-500 95.9% #35 of 70 Lab model card
GPQA Diamond 56.1% #145 of 229 Epoch AI run
AIME 2024 70.7% #61 of 88 Lab model card
OTIS Mock AIME 2024-2025 30% #147 of 212 Epoch AI run
AIME 2025 62.8% #124 of 143 Lab model card
ARC-AGI 5.00% #90 of 95 Official leaderboard
ARC-AGI-2 0% #98 of 98 Official leaderboard

Coding

BenchmarkScoreRankSource
LiveCodeBench v5 55.8% #34 of 47 Lab model card
LiveCodeBench v6 47.4% #84 of 90 Lab model card

Agents and tools

BenchmarkScoreRankSource
Galileo Agent Leaderboard v2 16% #21 of 21 Official leaderboard

Vision, audio and robotics

BenchmarkScoreRankSource
MMMU 66% #76 of 103 Lab model card
MathVista 66.8% #55 of 70 Lab model card
MMMU-Pro 51.8% #77 of 90 Lab model card

Its lab's numbers put it at 129.9, number 275 of 404

The Self-Reported Index fits one score to the 10 capability results Mistral AI reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 127.9 to 131.8. How we compute it.

Its lab's number next to a leaderboard

Where the benchmark's own leaderboard, or Epoch AI's run, also has this model. Gaps mostly come from different harnesses, effort levels and task subsets. Every pair.

Benchmark Lab says Board says Gap
GPQA Diamond 68.2% GPQA, SFT + RL (Magistral Small) column (Table 3) 56.1% Epoch AI run +12 pts

Its launch post

The benchmarks the lab chose to headline, with the scores it gave.

Its system card

Every result in the longer documents the lab published with it. 6 of them aren't in the launch post.