← All comparisons
GPT-5.2 (medium) vs Claude 4.1 Opus (Non-reasoning)
OpenAI vs Anthropic — side-by-side benchmark comparison
| GPT-5.2 (medium) | Claude 4.1 Opus (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 46.6 | 36.0 |
| Coding Index | 44.2 | — |
| Math Index | 96.7 | — |
| Output speed (tok/s) | 0.0 | 44.7 |
| Blended price ($/1M) | $4.81 | $32.81 |
| Time to first token (s) | 0.00s | 1.63s |
| aime | — | — |
| aime 25 | 96.7% | — |
| artificial analysis coding index | 44.20 | — |
| artificial analysis intelligence index | 46.60 | 36.00 |
| artificial analysis math index | 96.70 | — |
| gpqa | 86.4% | — |
| hle | 24.9% | — |
| ifbench | 65.2% | — |
| lcr | 63.3% | — |
| livecodebench | 89.4% | — |
| math 500 | — | — |
| mmlu pro | 85.9% | — |
| scicode | 46.2% | — |
| tau2 | 74.3% | — |
| terminalbench hard | 43.2% | — |
Benchmark data from Artificial Analysis.