← All comparisons
GPT-5 nano (medium) vs Claude 3.5 Sonnet (Oct '24)
OpenAI vs Anthropic — side-by-side benchmark comparison
| GPT-5 nano (medium) | Claude 3.5 Sonnet (Oct '24) | |
|---|---|---|
| Intelligence Index | 25.9 | 15.9 |
| Coding Index | 22.9 | 30.2 |
| Math Index | 78.3 | — |
| Output speed (tok/s) | 171.9 | 0.0 |
| Blended price ($/1M) | $0.14 | $6.56 |
| Time to first token (s) | 43.05s | 0.00s |
| aime | — | 15.7% |
| aime 25 | 78.3% | — |
| artificial analysis coding index | 22.90 | 30.20 |
| artificial analysis intelligence index | 25.90 | 15.90 |
| artificial analysis math index | 78.30 | — |
| gpqa | 67.0% | 59.9% |
| hle | 7.6% | 3.9% |
| ifbench | 65.9% | — |
| lcr | 40.0% | — |
| livecodebench | 76.3% | 38.1% |
| math 500 | — | 77.1% |
| mmlu pro | 77.2% | 77.2% |
| scicode | 33.8% | 36.6% |
| tau2 | 30.4% | — |
| terminalbench hard | 17.4% | — |
Benchmark data from Artificial Analysis.