← All comparisons
Gemini 2.5 Pro Preview (May' 25) vs Qwen3 235B A22B (Reasoning)
Google vs Alibaba — side-by-side benchmark comparison
| Gemini 2.5 Pro Preview (May' 25) | Qwen3 235B A22B (Reasoning) | |
|---|---|---|
| Intelligence Index | 29.5 | 19.8 |
| Coding Index | — | 17.4 |
| Math Index | — | 82.0 |
| Output speed (tok/s) | 0.0 | 58.3 |
| Blended price ($/1M) | $3.44 | $2.63 |
| Time to first token (s) | 0.00s | 1.37s |
| aime | 84.3% | 84.0% |
| aime 25 | — | 82.0% |
| artificial analysis coding index | — | 17.40 |
| artificial analysis intelligence index | 29.50 | 19.80 |
| artificial analysis math index | — | 82.00 |
| gpqa | 82.2% | 70.0% |
| hle | 15.4% | 11.7% |
| ifbench | — | 38.7% |
| lcr | — | 0.0% |
| livecodebench | 77.0% | 62.2% |
| math 500 | 98.6% | 93.0% |
| mmlu pro | 83.7% | 82.8% |
| scicode | 41.6% | 39.9% |
| tau2 | — | 24.0% |
| terminalbench hard | — | 6.1% |
Benchmark data from Artificial Analysis.