← All comparisons
Molmo2-8B vs Gemma 3 4B Instruct
Allen Institute for AI vs Google — side-by-side benchmark comparison
| Molmo2-8B | Gemma 3 4B Instruct | |
|---|---|---|
| Intelligence Index | 7.3 | 6.3 |
| Coding Index | 4.4 | 2.9 |
| Math Index | — | 12.7 |
| Output speed (tok/s) | 0.0 | 0.0 |
| Blended price ($/1M) | $0.00 | $0.05 |
| Time to first token (s) | 0.00s | 0.00s |
| aime | — | 6.3% |
| aime 25 | — | 12.7% |
| artificial analysis coding index | 4.40 | 2.90 |
| artificial analysis intelligence index | 7.30 | 6.30 |
| artificial analysis math index | — | 12.70 |
| gpqa | 42.5% | 29.1% |
| hle | 4.4% | 5.2% |
| ifbench | 26.9% | 28.3% |
| lcr | 0.0% | 5.7% |
| livecodebench | — | 11.2% |
| math 500 | — | 76.6% |
| mmlu pro | — | 41.7% |
| scicode | 13.3% | 7.3% |
| tau2 | 0.0% | 5.0% |
| terminalbench hard | 0.0% | 0.8% |
Benchmark data from Artificial Analysis.