← All benchmarks
Magistral Small 1.2
Mistral · Released 2025-09-17 · Independent benchmark scores from Artificial Analysis
Intelligence Index
8.6
Output speed
76.0
tokens/sec
Blended price
$0.75
per 1M tokens
Time to first token
0.81s
Model metadata
| Model ID | 70152cb0-fb36-4732-a925-89ef40994be1 |
|---|---|
| Slug | magistral-small-2509 |
| Release date | 2025-09-17 |
| Provider | Mistral |
| Provider slug | mistral |
| Output speed | 76.0 tok/s |
| Time to first token | 0.81s |
| Time to first answer | 27.14s |
| Blended price ($/1M) | $0.75 |
| Input price ($/1M) | $0.50 |
| Output price ($/1M) | $1.50 |
All evaluation scores (18)
| aime | — |
|---|---|
| aime 25 | 80.3% |
| artificial analysis coding index | 14.70 |
| artificial analysis intelligence index | 8.60 |
| artificial analysis math index | 80.30 |
| gpqa | 66.3% |
| hle | 6.4% |
| ifbench | 44.4% |
| lcr | 19.3% |
| livecodebench | 72.3% |
| math 500 | — |
| mmlu pro | 76.8% |
| scicode | — |
| tau banking | 4.5% |
| tau2 | 27.8% |
| terminalbench hard | 4.5% |
| terminalbench v2 1 | 4.5% |
| terminalbench v4 0 | — |
Source: artificialanalysis.ai/models/magistral-small-2509
Benchmark data from Artificial Analysis.