← All comparisons
Gemini 2.5 Flash-Lite (Non-reasoning) vs Claude 4.1 Opus (Reasoning)
Google vs Anthropic — side-by-side benchmark comparison
| Gemini 2.5 Flash-Lite (Non-reasoning) | Claude 4.1 Opus (Reasoning) | |
|---|---|---|
| Intelligence Index | 12.7 | 42.0 |
| Coding Index | 7.4 | 36.5 |
| Math Index | 35.3 | 80.3 |
| Output speed (tok/s) | 294.2 | 44.5 |
| Blended price ($/1M) | $0.17 | $32.81 |
| Time to first token (s) | 0.34s | 8.55s |
| aime | 50.0% | — |
| aime 25 | 35.3% | 80.3% |
| artificial analysis coding index | 7.40 | 36.50 |
| artificial analysis intelligence index | 12.70 | 42.00 |
| artificial analysis math index | 35.30 | 80.30 |
| gpqa | 47.4% | 80.9% |
| hle | 3.7% | 11.9% |
| ifbench | 31.5% | 55.4% |
| lcr | 31.3% | 66.3% |
| livecodebench | 40.0% | 65.4% |
| math 500 | 92.6% | — |
| mmlu pro | 72.4% | 88.0% |
| scicode | 17.7% | 40.9% |
| tau2 | 19.0% | 71.4% |
| terminalbench hard | 2.3% | 34.3% |
Benchmark data from Artificial Analysis.