← All comparisons
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) vs Claude 4.1 Opus (Non-reasoning)
Google vs Anthropic — side-by-side benchmark comparison
| Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) | Claude 4.1 Opus (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 21.6 | 36.0 |
| Coding Index | 18.2 | — |
| Math Index | 68.7 | — |
| Output speed (tok/s) | 0.0 | 44.7 |
| Blended price ($/1M) | $0.17 | $32.81 |
| Time to first token (s) | 0.00s | 1.63s |
| aime | — | — |
| aime 25 | 68.7% | — |
| artificial analysis coding index | 18.20 | — |
| artificial analysis intelligence index | 21.60 | 36.00 |
| artificial analysis math index | 68.70 | — |
| gpqa | 70.9% | — |
| hle | 6.6% | — |
| ifbench | 52.6% | — |
| lcr | 59.0% | — |
| livecodebench | 68.8% | — |
| math 500 | — | — |
| mmlu pro | 80.8% | — |
| scicode | 28.7% | — |
| tau2 | 30.7% | — |
| terminalbench hard | 12.9% | — |
Benchmark data from Artificial Analysis.