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
| Benchmark | Score | Rank |
|---|---|---|
| AIME 2024/2025 | 30.0% | #9 / 22 |
Reasoning
| Benchmark | Score | Rank |
|---|---|---|
| GPQA Diamond | 48.4% | #14 / 28 |
| ARC-AGI | 5.0% | #6 / 7 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.