← All comparisons
GPT-5.1 Codex mini (high) vs Claude 3.5 Sonnet (June '24)
OpenAI vs Anthropic — side-by-side benchmark comparison
| GPT-5.1 Codex mini (high) | Claude 3.5 Sonnet (June '24) | |
|---|---|---|
| Intelligence Index | 38.6 | 14.2 |
| Coding Index | 36.4 | 26.0 |
| Math Index | 91.7 | — |
| Output speed (tok/s) | 218.5 | 0.0 |
| Blended price ($/1M) | $0.69 | $6.56 |
| Time to first token (s) | 3.35s | 0.00s |
| aime | — | 9.7% |
| aime 25 | 91.7% | — |
| artificial analysis coding index | 36.40 | 26.00 |
| artificial analysis intelligence index | 38.60 | 14.20 |
| artificial analysis math index | 91.70 | — |
| gpqa | 81.3% | 56.0% |
| hle | 16.9% | 3.7% |
| ifbench | 67.9% | — |
| lcr | 62.7% | — |
| livecodebench | 83.6% | — |
| math 500 | — | 69.5% |
| mmlu pro | 82.0% | 75.1% |
| scicode | 42.6% | 31.6% |
| tau2 | 62.9% | — |
| terminalbench hard | 33.3% | — |
Benchmark data from Artificial Analysis.