← All comparisons
Gemini 2.0 Flash (experimental) vs Qwen3 4B (Non-reasoning)
Google vs Alibaba — side-by-side benchmark comparison
| Gemini 2.0 Flash (experimental) | Qwen3 4B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 16.8 | 12.5 |
| Coding Index | — | — |
| Math Index | — | — |
| Output speed (tok/s) | 0.0 | 103.5 |
| Blended price ($/1M) | $0.00 | $0.19 |
| Time to first token (s) | 0.00s | 1.02s |
| aime | 30.0% | 21.3% |
| aime 25 | — | — |
| artificial analysis coding index | — | — |
| artificial analysis intelligence index | 16.80 | 12.50 |
| artificial analysis math index | — | — |
| gpqa | 63.6% | 39.8% |
| hle | 4.7% | 3.7% |
| ifbench | — | — |
| lcr | — | — |
| livecodebench | 21.0% | 23.3% |
| math 500 | 91.1% | 84.3% |
| mmlu pro | 78.2% | 58.6% |
| scicode | 34.0% | 16.7% |
| tau2 | — | — |
| terminalbench hard | — | — |
Benchmark data from Artificial Analysis.