Magistral Small 1.2
| Released | 2025-09-18 |
|---|---|
| Price per 1M | $0.50 in · $1.50 out |
| Overall score | 39.4 · rank 196 |
| Context window | 131,072 tokens |
| Provider | Mistral |
| Weights | Open weights (Apache 2.0) |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 39.4 | 196 | 7 |
| Overall: Coding | 43.8 | 165 | 4 |
| Overall: Reasoning | 37.0 | 208 | 4 |
| Overall: Agentic work | 42.7 | 176 | 4 |
| Overall: Frontend | 43.8 | 151 | 3 |
| Coding | 44.6 | 159 | 2 |
| Reasoning | 39.4 | 199 | 2 |
| Math | 41.4 | 159 | 1 |
| Knowledge | 23.7 | 168 | 1 |
| Multimodal | 41.2 | 102 | 1 |
| Value vs price | −7.5 | 168 |
Evidence
Coding 44.6 ±3.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SciCode | 35.2% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 72.1% | Vals AI | 2026-09-01 |
Reasoning 39.4 ±3.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CritPt | 0.3% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 47.6% | Epoch AI | 2026-08-30 |
| GPQA Diamond (Vals run) | 58.3% | Vals AI | 2026-09-01 |
Math 41.4 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 28.1% | Epoch AI | 2026-08-30 |
Knowledge 23.7 ±4.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMLU-Pro (Vals run) | 62.1% | Vals AI | 2026-09-01 |
Multimodal 41.2 ±2.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMMU-Pro (Vals run) | 65.2% | Vals AI | 2026-09-01 |
Names matched to this model
magistral-small-1-2, magistral-small-2509