← All comparisons
Mistral Small 4 (Non-reasoning) vs Qwen3.6 27B (Non-reasoning)
Mistral vs Alibaba — side-by-side benchmark comparison
| Mistral Small 4 (Non-reasoning) | Qwen3.6 27B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 18.6 | 37.1 |
| Coding Index | 16.4 | 26.6 |
| Math Index | — | — |
| Output speed (tok/s) | 157.6 | 60.7 |
| Blended price ($/1M) | $0.26 | $1.35 |
| Time to first token (s) | 0.49s | 1.55s |
| aime | — | — |
| aime 25 | — | — |
| artificial analysis coding index | 16.40 | 26.60 |
| artificial analysis intelligence index | 18.60 | 37.10 |
| artificial analysis math index | — | — |
| gpqa | 57.1% | 82.9% |
| hle | 3.7% | 13.6% |
| ifbench | 32.8% | 45.7% |
| lcr | 21.3% | 55.0% |
| livecodebench | — | — |
| math 500 | — | — |
| mmlu pro | — | — |
| scicode | 28.1% | 37.3% |
| tau2 | 18.4% | 93.6% |
| terminalbench hard | 10.6% | 21.2% |
Benchmark data from Artificial Analysis.