← All comparisons
Gemini 2.5 Flash (Reasoning) vs Claude 4.1 Opus (Reasoning)
Google vs Anthropic — side-by-side benchmark comparison
| Gemini 2.5 Flash (Reasoning) | Claude 4.1 Opus (Reasoning) | |
|---|---|---|
| Intelligence Index | 27.0 | 42.0 |
| Coding Index | 22.2 | 36.5 |
| Math Index | 73.3 | 80.3 |
| Output speed (tok/s) | 205.5 | 44.5 |
| Blended price ($/1M) | $0.85 | $32.81 |
| Time to first token (s) | 10.67s | 8.55s |
| aime | 82.3% | — |
| aime 25 | 73.3% | 80.3% |
| artificial analysis coding index | 22.20 | 36.50 |
| artificial analysis intelligence index | 27.00 | 42.00 |
| artificial analysis math index | 73.30 | 80.30 |
| gpqa | 79.0% | 80.9% |
| hle | 11.1% | 11.9% |
| ifbench | 50.3% | 55.4% |
| lcr | 61.7% | 66.3% |
| livecodebench | 69.5% | 65.4% |
| math 500 | 98.1% | — |
| mmlu pro | 83.2% | 88.0% |
| scicode | 39.4% | 40.9% |
| tau2 | 31.6% | 71.4% |
| terminalbench hard | 13.6% | 34.3% |
Benchmark data from Artificial Analysis.