← All comparisons
Cogito v2.1 (Reasoning) vs Claude 4.1 Opus (Reasoning)
Deep Cogito vs Anthropic — side-by-side benchmark comparison
| Cogito v2.1 (Reasoning) | Claude 4.1 Opus (Reasoning) | |
|---|---|---|
| Intelligence Index | — | 42.0 |
| Coding Index | 24.8 | 36.5 |
| Math Index | 72.7 | 80.3 |
| Output speed (tok/s) | 80.7 | 44.5 |
| Blended price ($/1M) | $1.25 | $32.81 |
| Time to first token (s) | 0.51s | 8.55s |
| aime | — | — |
| aime 25 | 72.7% | 80.3% |
| artificial analysis coding index | 24.80 | 36.50 |
| artificial analysis intelligence index | — | 42.00 |
| artificial analysis math index | 72.70 | 80.30 |
| gpqa | 76.8% | 80.9% |
| hle | 11.0% | 11.9% |
| ifbench | 46.3% | 55.4% |
| lcr | 21.7% | 66.3% |
| livecodebench | 68.8% | 65.4% |
| math 500 | — | — |
| mmlu pro | 84.9% | 88.0% |
| scicode | 41.0% | 40.9% |
| tau2 | — | 71.4% |
| terminalbench hard | 16.7% | 34.3% |
Benchmark data from Artificial Analysis.