← All comparisons
DeepSeek V4 Pro (Reasoning, High Effort) vs OLMo 2 32B
DeepSeek vs Allen Institute for AI — side-by-side benchmark comparison
| DeepSeek V4 Pro (Reasoning, High Effort) | OLMo 2 32B | |
|---|---|---|
| Intelligence Index | 49.8 | 10.6 |
| Coding Index | 43.2 | 2.7 |
| Math Index | — | 3.3 |
| Output speed (tok/s) | 53.5 | 0.0 |
| Blended price ($/1M) | $0.54 | $0.00 |
| Time to first token (s) | 1.17s | 0.00s |
| aime | — | — |
| aime 25 | — | 3.3% |
| artificial analysis coding index | 43.20 | 2.70 |
| artificial analysis intelligence index | 49.80 | 10.60 |
| artificial analysis math index | — | 3.30 |
| gpqa | 90.5% | 32.8% |
| hle | 33.5% | 3.7% |
| ifbench | 71.3% | 38.1% |
| lcr | 65.0% | 0.0% |
| livecodebench | — | 6.8% |
| math 500 | — | — |
| mmlu pro | — | 51.1% |
| scicode | 46.4% | 8.0% |
| tau2 | 94.2% | 0.0% |
| terminalbench hard | 41.7% | 0.0% |
Benchmark data from Artificial Analysis.