← All comparisons
GPT-4o (Nov '24) vs Claude 3.5 Sonnet (Oct '24)
OpenAI vs Anthropic — side-by-side benchmark comparison
| GPT-4o (Nov '24) | Claude 3.5 Sonnet (Oct '24) | |
|---|---|---|
| Intelligence Index | 17.3 | 15.9 |
| Coding Index | 16.7 | 30.2 |
| Math Index | 6.0 | — |
| Output speed (tok/s) | 158.5 | 0.0 |
| Blended price ($/1M) | $4.38 | $6.56 |
| Time to first token (s) | 0.54s | 0.00s |
| aime | 15.0% | 15.7% |
| aime 25 | 6.0% | — |
| artificial analysis coding index | 16.70 | 30.20 |
| artificial analysis intelligence index | 17.30 | 15.90 |
| artificial analysis math index | 6.00 | — |
| gpqa | 54.3% | 59.9% |
| hle | 3.3% | 3.9% |
| ifbench | 34.3% | — |
| lcr | 0.0% | — |
| livecodebench | 30.9% | 38.1% |
| math 500 | 75.9% | 77.1% |
| mmlu pro | 74.8% | 77.2% |
| scicode | 33.3% | 36.6% |
| tau2 | 25.1% | — |
| terminalbench hard | 8.3% | — |
Benchmark data from Artificial Analysis.