← All comparisons
Qwen3 Omni 30B A3B (Reasoning) vs Qwen3 VL 8B Instruct
Alibaba vs Alibaba — side-by-side benchmark comparison
| Qwen3 Omni 30B A3B (Reasoning) | Qwen3 VL 8B Instruct | |
|---|---|---|
| Intelligence Index | 15.6 | 14.3 |
| Coding Index | 12.7 | 7.3 |
| Math Index | 74.0 | 27.3 |
| Output speed (tok/s) | 100.1 | 143.8 |
| Blended price ($/1M) | $0.43 | $0.31 |
| Time to first token (s) | 0.97s | 0.93s |
| aime | — | — |
| aime 25 | 74.0% | 27.3% |
| artificial analysis coding index | 12.70 | 7.30 |
| artificial analysis intelligence index | 15.60 | 14.30 |
| artificial analysis math index | 74.00 | 27.30 |
| gpqa | 72.6% | 42.7% |
| hle | 7.3% | 2.9% |
| ifbench | 43.4% | 32.3% |
| lcr | 0.0% | 15.3% |
| livecodebench | 67.9% | 33.2% |
| math 500 | — | — |
| mmlu pro | 79.2% | 68.6% |
| scicode | 30.6% | 17.4% |
| tau2 | 21.3% | 29.2% |
| terminalbench hard | 3.8% | 2.3% |
Benchmark data from Artificial Analysis.