Up to four models
Mistral Large 3 vs Magistral Small 1.2
On the index (GPT-4 = 100), Mistral Large 3 scores 117 and Magistral Small 1.2 117. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
11799–135
WeightsOpen
ReleasedDec 2025
Price in / out$0.50 / $1.5
Context262K
Core tests taken0 of 9
117101–133
WeightsOpen
ReleasedSep 2025
Price in / out–
Context–
Core tests taken1 of 9
Mistral Large 3 and Magistral Small 1.2 are too close to call (0 points apart, within the uncertainty).
Test by testthird-party results
Every result as a table
| Test | Mistral Large 3 | Magistral Small 1.2 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Vals AI) | 68.4% | 58.3% |
| MMLU Pro | 79.8% | 62.1% |
| MMMU Pro | 66.2% | 65.2% |
| Maths | ||
| MGSM | 85.4% | 86.3% |
| Coding | ||
| LiveCodeBench | 55.3% | 72.1% |
| Agents | ||
| DTBench | 65.1% | 61.3% |
| Professional work | ||
| CorpFin v2 | 61.0% | 44.0% |
| LegalBench | 79.1% | 40.0% |
| MortgageTax | 52.1% | 62.1% |
| TaxEval v2 | 73.1% | 60.3% |
Tests at least two of these models have taken. Bold rows are the core tests.