← All comparisons
gpt-oss-120b (high) vs Claude 4.5 Haiku (Reasoning)
OpenAI vs Anthropic — side-by-side benchmark comparison
| gpt-oss-120b (high) | Claude 4.5 Haiku (Reasoning) | |
|---|---|---|
| Intelligence Index | 33.3 | 37.1 |
| Coding Index | 28.6 | 32.6 |
| Math Index | 93.4 | 83.7 |
| Output speed (tok/s) | 356.8 | 142.2 |
| Blended price ($/1M) | $0.26 | $2.19 |
| Time to first token (s) | 0.51s | 10.48s |
| aime | — | — |
| aime 25 | 93.4% | 83.7% |
| artificial analysis coding index | 28.60 | 32.60 |
| artificial analysis intelligence index | 33.30 | 37.10 |
| artificial analysis math index | 93.40 | 83.70 |
| gpqa | 78.2% | 67.2% |
| hle | 18.5% | 9.7% |
| ifbench | 69.0% | 54.3% |
| lcr | 50.7% | 70.3% |
| livecodebench | 87.8% | 61.5% |
| math 500 | — | — |
| mmlu pro | 80.8% | 76.0% |
| scicode | 38.9% | 43.3% |
| tau2 | 65.8% | 54.7% |
| terminalbench hard | 23.5% | 27.3% |
Benchmark data from Artificial Analysis.