Magistral Small 2506 — Benchmarks

Benchmark scores for Magistral Small 2506 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Math

BenchmarkScoreOpen rankAll models
AIME 2024/202530.0%#15 / 34#103 / 155

Reasoning

BenchmarkScoreOpen rankAll models
GPQA Diamond48.4%#23 / 46#126 / 182
ARC-AGI5.0%#8 / 10#151 / 158

Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.