← All comparisons
Gemini 2.0 Pro Experimental (Feb '25) vs Qwen3 14B (Reasoning)
Google vs Alibaba — side-by-side benchmark comparison
| Gemini 2.0 Pro Experimental (Feb '25) | Qwen3 14B (Reasoning) | |
|---|---|---|
| Intelligence Index | 18.1 | 16.2 |
| Coding Index | 25.5 | 13.1 |
| Math Index | — | 55.7 |
| Output speed (tok/s) | 0.0 | 66.2 |
| Blended price ($/1M) | $0.00 | $0.73 |
| Time to first token (s) | 0.00s | 1.14s |
| aime | 36.0% | 76.3% |
| aime 25 | — | 55.7% |
| artificial analysis coding index | 25.50 | 13.10 |
| artificial analysis intelligence index | 18.10 | 16.20 |
| artificial analysis math index | — | 55.70 |
| gpqa | 62.2% | 60.4% |
| hle | 6.8% | 4.3% |
| ifbench | — | 40.5% |
| lcr | — | 0.0% |
| livecodebench | 34.7% | 52.3% |
| math 500 | 92.3% | 96.1% |
| mmlu pro | 80.5% | 77.4% |
| scicode | 31.3% | 31.6% |
| tau2 | — | 34.5% |
| terminalbench hard | — | 3.8% |
Benchmark data from Artificial Analysis.