Molmo-72B-0924: benchmark results
Provider: Allen AI. Access: Open.
Aggregate Rating 1421 ± 33, rank #5112 of 10916 rated models, from 32 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| KOFFVQA - Hallucination and Robustness | 70 | Score (%) | 58.6 |
| TurtleAI - Synthetic | 1.62 | Success rate (%) on the 619 TurtleAI-DSSyn tasks (generated | 47.1 |
| TurtleAI - Hand-Drawn | 4.9 | Success rate (%) on the 102 TurtleAI-DSCraft tasks (hand-dra | 41.2 |
| TurtleAI | 2.31 | Success rate (%) on all 823 TurtleAI-DS tasks (102 real, 102 | 38.2 |
| OddGridBench - Color | 25.5 | Accuracy (%) on the 200 images in which the odd item differs | 33.3 |
| OddGridBench - Rotation | 7 | Accuracy (%) on the 200 images in which the odd item differs | 33.3 |
| KOFFVQA - Relationship | 47.67 | Score (%) | 28.4 |
| OddGridBench - Two Attributes | 12 | Accuracy (%) on the 200 images in which the odd item differs | 27.8 |
| TurtleAI - Real | 3.92 | Success rate (%) on the 102 TurtleAI-DSReal tasks (XLogoOnli | 26.5 |
| KOFFVQA - Graph and Chart Understanding | 42.67 | Score (%) | 25.9 |
| OddGridBench - Four Attributes | 17.5 | Accuracy (%) on the 200 images in which the odd item differs | 25 |
| KOFFVQA - Table Understanding | 20.67 | Score (%) | 23.5 |
Interactive version: theaggregate.ai/model?slug=molmo-72b-0924 · How It Works · Data refreshed daily, snapshot 2026-11-08.