← All comparisons
Olmo 3.1 32B Think vs GPT-5.1 (Non-reasoning)
Allen Institute for AI vs OpenAI — side-by-side benchmark comparison
| Olmo 3.1 32B Think | GPT-5.1 (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 13.9 | 27.4 |
| Coding Index | 9.8 | 27.3 |
| Math Index | 77.3 | 38.0 |
| Output speed (tok/s) | 0.0 | 129.5 |
| Blended price ($/1M) | $0.00 | $3.44 |
| Time to first token (s) | 0.00s | 0.72s |
| aime | — | — |
| aime 25 | 77.3% | 38.0% |
| artificial analysis coding index | 9.80 | 27.30 |
| artificial analysis intelligence index | 13.90 | 27.40 |
| artificial analysis math index | 77.30 | 38.00 |
| gpqa | 59.1% | 64.3% |
| hle | 6.0% | 5.2% |
| ifbench | 66.0% | 43.2% |
| lcr | 0.0% | 44.0% |
| livecodebench | 69.5% | 49.4% |
| math 500 | — | — |
| mmlu pro | 76.3% | 80.1% |
| scicode | 29.3% | 36.5% |
| tau2 | 0.0% | 46.5% |
| terminalbench hard | 0.0% | 22.7% |
Benchmark data from Artificial Analysis.