← All comparisons
Mixtral 8x22B Instruct vs Qwen3 32B (Reasoning)
Mistral vs Alibaba — side-by-side benchmark comparison
| Mixtral 8x22B Instruct | Qwen3 32B (Reasoning) | |
|---|---|---|
| Intelligence Index | 9.8 | 16.5 |
| Coding Index | — | 13.8 |
| Math Index | — | 73.0 |
| Output speed (tok/s) | 0.0 | 93.4 |
| Blended price ($/1M) | $0.00 | $0.28 |
| Time to first token (s) | 0.00s | 1.13s |
| aime | 0.0% | 80.7% |
| aime 25 | — | 73.0% |
| artificial analysis coding index | — | 13.80 |
| artificial analysis intelligence index | 9.80 | 16.50 |
| artificial analysis math index | — | 73.00 |
| gpqa | 33.2% | 66.8% |
| hle | 4.1% | 8.3% |
| ifbench | — | 36.3% |
| lcr | — | 0.0% |
| livecodebench | 14.8% | 54.6% |
| math 500 | 54.5% | 96.1% |
| mmlu pro | 53.7% | 79.8% |
| scicode | 18.8% | 35.4% |
| tau2 | — | 29.8% |
| terminalbench hard | — | 3.0% |
Benchmark data from Artificial Analysis.