← All comparisons
Kimi K2.6 (Non-reasoning) vs Qwen3 VL 30B A3B Instruct
Kimi vs Alibaba — side-by-side benchmark comparison
| Kimi K2.6 (Non-reasoning) | Qwen3 VL 30B A3B Instruct | |
|---|---|---|
| Intelligence Index | 42.9 | 16.0 |
| Coding Index | 38.4 | 14.3 |
| Math Index | — | 72.3 |
| Output speed (tok/s) | 39.0 | 123.5 |
| Blended price ($/1M) | $1.71 | $0.30 |
| Time to first token (s) | 1.28s | 1.07s |
| aime | — | — |
| aime 25 | — | 72.3% |
| artificial analysis coding index | 38.40 | 14.30 |
| artificial analysis intelligence index | 42.90 | 16.00 |
| artificial analysis math index | — | 72.30 |
| gpqa | 78.8% | 69.5% |
| hle | 18.2% | 6.4% |
| ifbench | 44.3% | 33.1% |
| lcr | 57.7% | 23.7% |
| livecodebench | — | 47.6% |
| math 500 | — | — |
| mmlu pro | — | 76.4% |
| scicode | 39.5% | 30.8% |
| tau2 | 93.9% | 19.0% |
| terminalbench hard | 37.9% | 6.1% |
Benchmark data from Artificial Analysis.