← All comparisons
Olmo 3 7B Think vs DeepSeek V3.2 Exp (Reasoning)
Allen Institute for AI vs DeepSeek — side-by-side benchmark comparison
| Olmo 3 7B Think | DeepSeek V3.2 Exp (Reasoning) | |
|---|---|---|
| Intelligence Index | 9.4 | 32.9 |
| Coding Index | 7.6 | 33.3 |
| Math Index | 70.7 | 87.7 |
| Output speed (tok/s) | 0.0 | 0.0 |
| Blended price ($/1M) | $0.00 | $0.31 |
| Time to first token (s) | 0.00s | 0.00s |
| aime | — | — |
| aime 25 | 70.7% | 87.7% |
| artificial analysis coding index | 7.60 | 33.30 |
| artificial analysis intelligence index | 9.40 | 32.90 |
| artificial analysis math index | 70.70 | 87.70 |
| gpqa | 51.6% | 79.7% |
| hle | 5.7% | 13.8% |
| ifbench | 41.5% | 54.1% |
| lcr | 0.0% | 69.0% |
| livecodebench | 61.7% | 78.9% |
| math 500 | — | — |
| mmlu pro | 65.5% | 85.0% |
| scicode | 21.2% | 37.7% |
| tau2 | 0.0% | 33.9% |
| terminalbench hard | 0.8% | 31.1% |
Benchmark data from Artificial Analysis.