Long Video Understanding on LongVideoBench 2024 (val)
66.7AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4oSize=-, Frames=0.5 fps / 3842025.08 | 66.7 | |
| Gemini-1.5-ProSize=-, Frames=0.5 fps2025.08 | 64 | |
| Video-MTRSize=7B, Frames=802025.08 | 57.1 | |
| LLaVA-OneVisionSize=7B, Frames=322025.08 | 56.3 | |
| Qwen2.5-VL-7BSize=7B, Frames=7682025.08 | 56 | |
| Video-MTRSize=7B, Frames=642025.08 | 54.8 | |
| Video-MTRSize=7B, Frames=322025.08 | 52.3 | |
| Video-XLSize=7B, Frames=128 / 2562025.08 | 50.7 | |
| Qwen2.5-VL-7BSize=7B, Frames=802025.08 | 48.4 | |
| LongVASize=7B, Frames=128 / 2562025.08 | 47.8 | |
| Qwen2.5-VL-7BSize=7B, Frames=642025.08 | 47 | |
| Qwen2.5-VL-7BSize=7B, Frames=322025.08 | 45.8 | |
| Video-CCAMSize=9B, Frames=962025.08 | 43.1 | |
| VideoChat2Size=7B, Frames=162025.08 | 39.3 | |
| Video-LLaVASize=7B, Frames=82025.08 | 39.1 |