Long Video Understanding on LongVideoBench (LVB) (val)
66.7AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4o# Frame=384, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 66.7 | |
| Gemini-1.5-Pro# Frame=1 fps, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 64 | |
| Reflect-R1# Frame=768, Paradigm=Tool-Augmented Reasoning, Adaptive keyframe retrieval=false2026.06 | 62.5 | |
| TimeSearch-R-7B# Frame=768, Paradigm=Tool-Augmented Reasoning, Adaptive keyframe retrieval=false2026.06 | 61.2 | |
| Video-R1-7B# Frame=768, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 60.1 | |
| VL-Rethinker# Frame=768, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 55.58 | |
| Qwen2.5-VL-7B-Instruct# Frame=768, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 53.9 | |
| Vision-R1-7B# Frame=768, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 52.8 |