← All comparisons
GPT-5.1 Codex (high) vs Claude 4.1 Opus (Reasoning)
OpenAI vs Anthropic — side-by-side benchmark comparison
| GPT-5.1 Codex (high) | Claude 4.1 Opus (Reasoning) | |
|---|---|---|
| Intelligence Index | 43.1 | 42.0 |
| Coding Index | 36.6 | 36.5 |
| Math Index | 95.7 | 80.3 |
| Output speed (tok/s) | 182.1 | 44.5 |
| Blended price ($/1M) | $3.44 | $32.81 |
| Time to first token (s) | 5.42s | 8.55s |
| aime | — | — |
| aime 25 | 95.7% | 80.3% |
| artificial analysis coding index | 36.60 | 36.50 |
| artificial analysis intelligence index | 43.10 | 42.00 |
| artificial analysis math index | 95.70 | 80.30 |
| gpqa | 86.0% | 80.9% |
| hle | 23.4% | 11.9% |
| ifbench | 70.0% | 55.4% |
| lcr | 67.3% | 66.3% |
| livecodebench | 84.9% | 65.4% |
| math 500 | — | — |
| mmlu pro | 86.0% | 88.0% |
| scicode | 40.2% | 40.9% |
| tau2 | 83.0% | 71.4% |
| terminalbench hard | 34.8% | 34.3% |
Benchmark data from Artificial Analysis.