← All comparisons
Gemini 3.5 Flash (high) vs Molmo2-8B
Google vs Allen Institute for AI — side-by-side benchmark comparison
| Gemini 3.5 Flash (high) | Molmo2-8B | |
|---|---|---|
| Intelligence Index | 55.3 | 7.3 |
| Coding Index | 45.0 | 4.4 |
| Math Index | — | — |
| Output speed (tok/s) | 219.4 | 0.0 |
| Blended price ($/1M) | $3.38 | $0.00 |
| Time to first token (s) | 15.64s | 0.00s |
| aime | — | — |
| aime 25 | — | — |
| artificial analysis coding index | 45.00 | 4.40 |
| artificial analysis intelligence index | 55.30 | 7.30 |
| artificial analysis math index | — | — |
| gpqa | 92.2% | 42.5% |
| hle | 41.0% | 4.4% |
| ifbench | 76.3% | 26.9% |
| lcr | 69.3% | 0.0% |
| livecodebench | — | — |
| math 500 | — | — |
| mmlu pro | — | — |
| scicode | 53.1% | 13.3% |
| tau2 | 95.3% | 0.0% |
| terminalbench hard | 40.9% | 0.0% |
Benchmark data from Artificial Analysis.