← All comparisons
Qwen3 Omni 30B A3B Instruct vs Qwen3 235B A22B (Reasoning)
Alibaba vs Alibaba — side-by-side benchmark comparison
| Qwen3 Omni 30B A3B Instruct | Qwen3 235B A22B (Reasoning) | |
|---|---|---|
| Intelligence Index | 10.7 | 19.8 |
| Coding Index | 7.2 | 17.4 |
| Math Index | 52.3 | 82.0 |
| Output speed (tok/s) | 108.9 | 58.3 |
| Blended price ($/1M) | $0.43 | $2.63 |
| Time to first token (s) | 1.04s | 1.37s |
| aime | — | 84.0% |
| aime 25 | 52.3% | 82.0% |
| artificial analysis coding index | 7.20 | 17.40 |
| artificial analysis intelligence index | 10.70 | 19.80 |
| artificial analysis math index | 52.30 | 82.00 |
| gpqa | 62.0% | 70.0% |
| hle | 5.1% | 11.7% |
| ifbench | 31.2% | 38.7% |
| lcr | 0.0% | 0.0% |
| livecodebench | 42.2% | 62.2% |
| math 500 | — | 93.0% |
| mmlu pro | 72.5% | 82.8% |
| scicode | 18.6% | 39.9% |
| tau2 | 16.4% | 24.0% |
| terminalbench hard | 1.5% | 6.1% |
Benchmark data from Artificial Analysis.