- AIME24 pass@1 91.8
- AIME25 pass@1 83.5
- GPQA Diamond 76.3
- Livecodebench (v5) 75
Magistral Medium 1.2
Magistral Medium 1.2 places in the top three on LiveCodeBench v5.
11
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Mistral AI Sep 2025 Magistral Medium 1.2
Closed weights
No Epoch Capabilities Index score yet
- LiveCodeBench v5 75% #2 /47
- CorpFin v2 47.4% #99 /119
- TaxEval v2 61.9% #109 /129
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 | |
|---|---|---|---|---|
| SnorkelSpatial | 44.2% | #8 of 27 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2024 | 91.8% | #10 of 88 | Lab model card | |
| GPQA Diamond | 76.3% | #140 of 302 | Lab model card | |
| AIME 2025 | 83.5% | #78 of 143 | Lab model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v5 | 75% | #2 of 47 | Lab model card | |
| IOI (Vals) | 0.67% | #53 of 55 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vals Finance Agent v1 | 18.2% | #35 of 47 | Official leaderboard | |
| CorpFin v2 | 47.4% | #99 of 119 | Official leaderboard | |
| TaxEval v2 | 61.9% | #109 of 129 | Official leaderboard | |
| MortgageTax | 51.9% | #76 of 87 | Official leaderboard | |
| LegalBench | 54.8% | #132 of 138 | Official leaderboard |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.