Molmo-72B: benchmark results
Provider: Allen AI. Released 2024-09-24. Access: Open.
Aggregate Rating 1438 ± 38, rank #4256 of 10916 rated models, from 196 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| GQA | 86.67 | Overall Accuracy (%) | 98.8 |
| OpenVLM MMStar - Fine-Grained Perception | 62.8 | Accuracy (%) | 96.8 |
| OpenVLM MMMU - Clinical Medicine | 73.3 | Accuracy (%) | 94.5 |
| OpenVLM AI2D - Eclipses | 96.8 | Accuracy (%) | 93.1 |
| OpenVLM AI2D - Volcano | 100 | Accuracy (%) | 92.8 |
| OpenVLM RealWorldQA | 73.7 | Accuracy (%) | 92.7 |
| OpenVLM MMBench V1.1 CN - Physical Relation | 68 | Accuracy (%) | 90.1 |
| OpenVLM MMStar - Coarse Perception | 72.4 | Accuracy (%) | 89.9 |
| OpenVLM MMBench V1.1 CN - Spatial Relationship | 73.3 | Accuracy (%) | 86.2 |
| OpenVLM MME - Count | 171.7 | Score | 84.8 |
| OpenVLM MME - Posters | 182.3 | Score | 84.6 |
| OpenVLM MMMU - Management | 66.7 | Accuracy (%) | 84.5 |
Interactive version: theaggregate.ai/model?slug=molmo-72b · How It Works · Data refreshed daily, snapshot 2026-11-08.