← All comparisons
o4-mini (high) vs Claude 3.5 Sonnet (Oct '24)
OpenAI vs Anthropic — side-by-side benchmark comparison
| o4-mini (high) | Claude 3.5 Sonnet (Oct '24) | |
|---|---|---|
| Intelligence Index | 33.1 | 15.9 |
| Coding Index | 25.6 | 30.2 |
| Math Index | 90.7 | — |
| Output speed (tok/s) | 160.5 | 0.0 |
| Blended price ($/1M) | $1.93 | $6.56 |
| Time to first token (s) | 23.07s | 0.00s |
| aime | 94.0% | 15.7% |
| aime 25 | 90.7% | — |
| artificial analysis coding index | 25.60 | 30.20 |
| artificial analysis intelligence index | 33.10 | 15.90 |
| artificial analysis math index | 90.70 | — |
| gpqa | 78.4% | 59.9% |
| hle | 17.5% | 3.9% |
| ifbench | 68.7% | — |
| lcr | 55.0% | — |
| livecodebench | 85.9% | 38.1% |
| math 500 | 98.9% | 77.1% |
| mmlu pro | 83.2% | 77.2% |
| scicode | 46.5% | 36.6% |
| tau2 | 55.6% | — |
| terminalbench hard | 15.2% | — |
Benchmark data from Artificial Analysis.