← All comparisons
Qwen3 Omni 30B A3B (Reasoning) vs DeepSeek V3.2 Exp (Reasoning)
Alibaba vs DeepSeek — side-by-side benchmark comparison
| Qwen3 Omni 30B A3B (Reasoning) | DeepSeek V3.2 Exp (Reasoning) | |
|---|---|---|
| Intelligence Index | 15.6 | 32.9 |
| Coding Index | 12.7 | 33.3 |
| Math Index | 74.0 | 87.7 |
| Output speed (tok/s) | 100.1 | 0.0 |
| Blended price ($/1M) | $0.43 | $0.31 |
| Time to first token (s) | 0.97s | 0.00s |
| aime | — | — |
| aime 25 | 74.0% | 87.7% |
| artificial analysis coding index | 12.70 | 33.30 |
| artificial analysis intelligence index | 15.60 | 32.90 |
| artificial analysis math index | 74.00 | 87.70 |
| gpqa | 72.6% | 79.7% |
| hle | 7.3% | 13.8% |
| ifbench | 43.4% | 54.1% |
| lcr | 0.0% | 69.0% |
| livecodebench | 67.9% | 78.9% |
| math 500 | — | — |
| mmlu pro | 79.2% | 85.0% |
| scicode | 30.6% | 37.7% |
| tau2 | 21.3% | 33.9% |
| terminalbench hard | 3.8% | 31.1% |
Benchmark data from Artificial Analysis.