- AIME24 pass@1 86.1
- AIME25 pass@1 77.3
- GPQA Diamond 70.1
- Livecodebench (v5) 70.9
Magistral Small 1.2
Magistral Small 1.2.
11
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Mistral AI Sep 2025 Magistral Small 1.2
Open weights
No Epoch Capabilities Index score yet
- GPQA Diamond 47.6% #165 /229
- LiveCodeBench v5 70.9% #9 /47
- CorpFin v2 44% #107 /119
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
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AraGen | 32.3% | #29 of 42 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2024 | 86.1% | #29 of 88 | Lab model card | |
| AIME 2025 | 77.3% | #98 of 143 | Lab model card | |
| OTIS Mock AIME 2024-2025 | 28.1% | #149 of 212 | Epoch AI run | |
| GPQA Diamond | 47.6% | #165 of 229 | Epoch AI run | |
| DTBench | 61.3% | #140 of 180 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v5 | 70.9% | #9 of 47 | Lab model card |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MortgageTax | 62.1% | #54 of 87 | Official leaderboard | |
| TaxEval v2 | 60.3% | #115 of 129 | Official leaderboard | |
| CorpFin v2 | 44% | #107 of 119 | Official leaderboard | |
| LegalBench | 40% | #137 of 138 | Official leaderboard |
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 | 70.1% GPQA Diamond | 47.6% Epoch AI run | +22 pts |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.