← All comparisons
Claude 4.5 Haiku (Reasoning) vs Qwen3 0.6B (Reasoning)
Anthropic vs Alibaba — side-by-side benchmark comparison
| Claude 4.5 Haiku (Reasoning) | Qwen3 0.6B (Reasoning) | |
|---|---|---|
| Intelligence Index | 37.1 | 6.5 |
| Coding Index | 32.6 | 0.9 |
| Math Index | 83.7 | 18.0 |
| Output speed (tok/s) | 142.2 | 222.0 |
| Blended price ($/1M) | $2.19 | $0.40 |
| Time to first token (s) | 10.48s | 1.07s |
| aime | — | 10.0% |
| aime 25 | 83.7% | 18.0% |
| artificial analysis coding index | 32.60 | 90.0% |
| artificial analysis intelligence index | 37.10 | 6.50 |
| artificial analysis math index | 83.70 | 18.00 |
| gpqa | 67.2% | 23.9% |
| hle | 9.7% | 5.7% |
| ifbench | 54.3% | 23.3% |
| lcr | 70.3% | 0.0% |
| livecodebench | 61.5% | 12.1% |
| math 500 | — | 75.0% |
| mmlu pro | 76.0% | 34.7% |
| scicode | 43.3% | 2.8% |
| tau2 | 54.7% | 21.1% |
| terminalbench hard | 27.3% | 0.0% |
Benchmark data from Artificial Analysis.