← All comparisons
Molmo2-8B vs Llama 3.1 Instruct 8B
Allen Institute for AI vs Meta — side-by-side benchmark comparison
| Molmo2-8B | Llama 3.1 Instruct 8B | |
|---|---|---|
| Intelligence Index | 7.3 | 11.8 |
| Coding Index | 4.4 | 4.9 |
| Math Index | — | 4.3 |
| Output speed (tok/s) | 0.0 | 212.7 |
| Blended price ($/1M) | $0.00 | $0.10 |
| Time to first token (s) | 0.00s | 0.45s |
| aime | — | 7.7% |
| aime 25 | — | 4.3% |
| artificial analysis coding index | 4.40 | 4.90 |
| artificial analysis intelligence index | 7.30 | 11.80 |
| artificial analysis math index | — | 4.30 |
| gpqa | 42.5% | 25.9% |
| hle | 4.4% | 5.1% |
| ifbench | 26.9% | 28.6% |
| lcr | 0.0% | 15.7% |
| livecodebench | — | 11.6% |
| math 500 | — | 51.9% |
| mmlu pro | — | 47.6% |
| scicode | 13.3% | 13.2% |
| tau2 | 0.0% | 16.4% |
| terminalbench hard | 0.0% | 0.8% |
Benchmark data from Artificial Analysis.