← All comparisons
Olmo 3.1 32B Instruct vs Llama 3.2 Instruct 3B
Allen Institute for AI vs Meta — side-by-side benchmark comparison
| Olmo 3.1 32B Instruct | Llama 3.2 Instruct 3B | |
|---|---|---|
| Intelligence Index | 12.2 | 9.7 |
| Coding Index | 5.6 | — |
| Math Index | — | 3.3 |
| Output speed (tok/s) | 0.0 | 52.7 |
| Blended price ($/1M) | $0.00 | $0.15 |
| Time to first token (s) | 0.00s | 0.65s |
| aime | — | 6.7% |
| aime 25 | — | 3.3% |
| artificial analysis coding index | 5.60 | — |
| artificial analysis intelligence index | 12.20 | 9.70 |
| artificial analysis math index | — | 3.30 |
| gpqa | 53.9% | 25.5% |
| hle | 4.9% | 5.2% |
| ifbench | 39.2% | 26.2% |
| lcr | 0.0% | 2.0% |
| livecodebench | — | 8.3% |
| math 500 | — | 48.9% |
| mmlu pro | — | 34.7% |
| scicode | 16.7% | 5.2% |
| tau2 | 21.3% | 21.1% |
| terminalbench hard | 0.0% | — |
Benchmark data from Artificial Analysis.