Video Multi-modal Evaluation on VideoMME (test)
64.4Overall ScoreLLaVA-Video-7B
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| LLaVA-Video-7BBackbone=LLaVA-Video-7B, Retention Ratio=100%, Input Frames=642026.03 | 64.4 | 77.3 | 62.4 | 53.4 | |
| HoliTomBackbone=LLaVA-Video-7B, Retention Ratio=25%, Input Frames=642026.03 | 63 | 74.6 | 62.3 | 52.1 | |
| V-CASTBackbone=LLaVA-Video-7B, Retention Ratio=25%, Input Frames=642026.03 | 62.7 | 74 | 61.2 | 52.4 | |
| VisionZipBackbone=LLaVA-Video-7B, Retention Ratio=25%, Input Frames=642026.03 | 62.5 | 73.6 | 62.3 | 51.9 | |
| VidCom2Backbone=LLaVA-Video-7B, Retention Ratio=25%, Input Frames=642026.03 | 61.7 | 73 | 61.7 | 50 | |
| SparseVLMBackbone=LLaVA-Video-7B, Retention Ratio=25%, Input Frames=642026.03 | 60.1 | 71.1 | 59.1 | 50.1 | |
| FastVBackbone=LLaVA-Video-7B, Retention Ratio=25%, Input Frames=642026.03 | 58.6 | 68.7 | 58.4 | 48.7 |