| Benchmark | Model | Rank | Results |
|---|---|---|---|
| multimodal-reasoning-on-visulogic | UVU 3B | – | Accuracy: 26.4 |
| science-question-answering-on-scienceqa | UVU 3B | – | Avg. Accuracy: 91.3 |
| vision-language-understanding-on-mmstar | UVU 3B | – | Accuracy: 55.0 |
| visual-question-answering-vqa-on-blink | UVU 3B | – | Accuracy: 52.8 |