← All comparisons
Qwen3.6 27B (Reasoning) vs Qwen3 30B A3B (Reasoning)
Alibaba vs Alibaba — side-by-side benchmark comparison
| Qwen3.6 27B (Reasoning) | Qwen3 30B A3B (Reasoning) | |
|---|---|---|
| Intelligence Index | 45.8 | 15.3 |
| Coding Index | 36.5 | 11.0 |
| Math Index | — | 72.3 |
| Output speed (tok/s) | 57.4 | 64.1 |
| Blended price ($/1M) | $1.35 | $0.18 |
| Time to first token (s) | 1.46s | 1.18s |
| aime | — | 75.3% |
| aime 25 | — | 72.3% |
| artificial analysis coding index | 36.50 | 11.00 |
| artificial analysis intelligence index | 45.80 | 15.30 |
| artificial analysis math index | — | 72.30 |
| gpqa | 84.2% | 61.6% |
| hle | 21.6% | 6.6% |
| ifbench | 67.6% | 41.5% |
| lcr | 68.7% | 0.0% |
| livecodebench | — | 50.6% |
| math 500 | — | 95.9% |
| mmlu pro | — | 77.7% |
| scicode | 39.8% | 28.5% |
| tau2 | 94.2% | 26.0% |
| terminalbench hard | 34.8% | 2.3% |
Benchmark data from Artificial Analysis.