← All comparisons
GPT-5 (minimal) vs Claude 3.5 Sonnet (Oct '24)
OpenAI vs Anthropic — side-by-side benchmark comparison
| GPT-5 (minimal) | Claude 3.5 Sonnet (Oct '24) | |
|---|---|---|
| Intelligence Index | 23.9 | 15.9 |
| Coding Index | 25.0 | 30.2 |
| Math Index | 31.7 | — |
| Output speed (tok/s) | 76.8 | 0.0 |
| Blended price ($/1M) | $3.44 | $6.56 |
| Time to first token (s) | 0.84s | 0.00s |
| aime | 36.7% | 15.7% |
| aime 25 | 31.7% | — |
| artificial analysis coding index | 25.00 | 30.20 |
| artificial analysis intelligence index | 23.90 | 15.90 |
| artificial analysis math index | 31.70 | — |
| gpqa | 67.3% | 59.9% |
| hle | 5.4% | 3.9% |
| ifbench | 45.6% | — |
| lcr | 25.0% | — |
| livecodebench | 55.8% | 38.1% |
| math 500 | 86.1% | 77.1% |
| mmlu pro | 80.6% | 77.2% |
| scicode | 38.8% | 36.6% |
| tau2 | 67.0% | — |
| terminalbench hard | 18.2% | — |
Benchmark data from Artificial Analysis.