← All benchmarks
Mistral Small 4 (Reasoning)
Mistral · Released 2026-03-16 · Independent benchmark scores from Artificial Analysis
Intelligence Index
19.6
Output speed
177.3
tokens/sec
Blended price
$0.26
per 1M tokens
Time to first token
0.56s
Model metadata
| Model ID | 3fd96175-4ef1-434c-8795-f873aec2abc1 |
|---|---|
| Slug | mistral-small-4 |
| Release date | 2026-03-16 |
| Provider | Mistral |
| Provider slug | mistral |
| Output speed | 177.3 tok/s |
| Time to first token | 0.56s |
| Time to first answer | 11.85s |
| Blended price ($/1M) | $0.26 |
| Input price ($/1M) | $0.15 |
| Output price ($/1M) | $0.60 |
All evaluation scores (17)
| aime | — |
|---|---|
| aime 25 | — |
| artificial analysis coding index | 26.60 |
| artificial analysis intelligence index | 19.60 |
| artificial analysis math index | — |
| gpqa | 76.9% |
| hle | 9.5% |
| ifbench | 48.2% |
| lcr | 44.7% |
| livecodebench | — |
| math 500 | — |
| mmlu pro | — |
| scicode | 38.0% |
| tau banking | 5.2% |
| tau2 | 41.2% |
| terminalbench hard | 17.4% |
| terminalbench v2 1 | 21.0% |
Source: artificialanalysis.ai/models/mistral-small-4
Benchmark data from Artificial Analysis.