← All comparisons
Olmo 3 7B Instruct vs Qwen3 32B (Non-reasoning)
Allen Institute for AI vs Alibaba — side-by-side benchmark comparison
| Olmo 3 7B Instruct | Qwen3 32B (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 8.1 | 14.5 |
| Coding Index | 3.4 | — |
| Math Index | 41.3 | 19.7 |
| Output speed (tok/s) | 0.0 | 94.2 |
| Blended price ($/1M) | $0.13 | $0.26 |
| Time to first token (s) | 0.00s | 1.12s |
| aime | — | 30.3% |
| aime 25 | 41.3% | 19.7% |
| artificial analysis coding index | 3.40 | — |
| artificial analysis intelligence index | 8.10 | 14.50 |
| artificial analysis math index | 41.30 | 19.70 |
| gpqa | 40.0% | 53.5% |
| hle | 5.8% | 4.3% |
| ifbench | 32.8% | 31.5% |
| lcr | 0.0% | 0.0% |
| livecodebench | 26.6% | 28.8% |
| math 500 | — | 86.9% |
| mmlu pro | 52.2% | 72.7% |
| scicode | 10.3% | 28.0% |
| tau2 | 12.6% | — |
| terminalbench hard | 0.0% | — |
Benchmark data from Artificial Analysis.