← All comparisons
Llama 3.2 Instruct 11B (Vision) vs Molmo2-8B
Meta vs Allen Institute for AI — side-by-side benchmark comparison
| Llama 3.2 Instruct 11B (Vision) | Molmo2-8B | |
|---|---|---|
| Intelligence Index | 8.7 | 7.3 |
| Coding Index | 4.2 | 4.4 |
| Math Index | 1.7 | — |
| Output speed (tok/s) | 77.0 | 0.0 |
| Blended price ($/1M) | $0.24 | $0.00 |
| Time to first token (s) | 0.49s | 0.00s |
| aime | 9.3% | — |
| aime 25 | 1.7% | — |
| artificial analysis coding index | 4.20 | 4.40 |
| artificial analysis intelligence index | 8.70 | 7.30 |
| artificial analysis math index | 1.70 | — |
| gpqa | 22.1% | 42.5% |
| hle | 5.2% | 4.4% |
| ifbench | 30.4% | 26.9% |
| lcr | 11.7% | 0.0% |
| livecodebench | 11.0% | — |
| math 500 | 51.6% | — |
| mmlu pro | 46.4% | — |
| scicode | 11.2% | 13.3% |
| tau2 | 14.6% | 0.0% |
| terminalbench hard | 0.8% | 0.0% |
Benchmark data from Artificial Analysis.