| 3d-question-answering-3d-qa-on-sqa3d | LLaVA-Video | – | Exact Match: 48.5 |
| video-question-answering-on-implicitqa | LLaVA-Video - 7B | – | Average Accuracy: 42.1Macro Average Accuracy: 46.3 |
| video-question-answering-on-next-qa | LLaVA-Video | – | Accuracy: 83.2 |
| video-question-answering-on-tvbench | LLaVA-Video 72B | – | Average Accuracy: 50.0 |
| video-question-answering-on-tvbench | LLaVA-Video 7B | – | Average Accuracy: 45.6 |
| visual-question-answering-vqa-on-vlm2-bench | LLaVA-Video-7B | – | GC-mat: 18.53GC-trk: 12.79OC-cpr: 54.72OC-cnt: 62.47… |
| zero-shot-video-question-answer-on-longvideobench | LLaVA-Video | – | Accuracy (% ): 61.9 |