← All comparisons
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) vs Claude 4.1 Opus (Reasoning)
Google vs Anthropic — side-by-side benchmark comparison
| Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) | Claude 4.1 Opus (Reasoning) | |
|---|---|---|
| Intelligence Index | 19.4 | 42.0 |
| Coding Index | 14.5 | 36.5 |
| Math Index | 46.7 | 80.3 |
| Output speed (tok/s) | 0.0 | 44.5 |
| Blended price ($/1M) | $0.17 | $32.81 |
| Time to first token (s) | 0.00s | 8.55s |
| aime | — | — |
| aime 25 | 46.7% | 80.3% |
| artificial analysis coding index | 14.50 | 36.50 |
| artificial analysis intelligence index | 19.40 | 42.00 |
| artificial analysis math index | 46.70 | 80.30 |
| gpqa | 65.1% | 80.9% |
| hle | 4.6% | 11.9% |
| ifbench | 41.8% | 55.4% |
| lcr | 48.0% | 66.3% |
| livecodebench | 64.1% | 65.4% |
| math 500 | — | — |
| mmlu pro | 79.6% | 88.0% |
| scicode | 28.5% | 40.9% |
| tau2 | 30.4% | 71.4% |
| terminalbench hard | 7.6% | 34.3% |
Benchmark data from Artificial Analysis.