← All comparisons
GPT-5.2 (Non-reasoning) vs Qwen3 0.6B (Non-reasoning)
OpenAI vs Alibaba — side-by-side benchmark comparison
| GPT-5.2 (Non-reasoning) | Qwen3 0.6B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 33.6 | 5.7 |
| Coding Index | 34.7 | 1.4 |
| Math Index | 51.0 | 10.3 |
| Output speed (tok/s) | 72.8 | 221.3 |
| Blended price ($/1M) | $4.81 | $0.19 |
| Time to first token (s) | 0.61s | 0.90s |
| aime | — | 1.7% |
| aime 25 | 51.0% | 10.3% |
| artificial analysis coding index | 34.70 | 1.40 |
| artificial analysis intelligence index | 33.60 | 5.70 |
| artificial analysis math index | 51.00 | 10.30 |
| gpqa | 71.2% | 23.1% |
| hle | 7.3% | 5.2% |
| ifbench | 47.4% | 21.9% |
| lcr | 38.0% | 0.0% |
| livecodebench | 66.9% | 7.3% |
| math 500 | — | 52.1% |
| mmlu pro | 81.4% | 23.1% |
| scicode | 40.4% | 4.1% |
| tau2 | 46.5% | 14.6% |
| terminalbench hard | 31.8% | 0.0% |
Benchmark data from Artificial Analysis.