Offline Video Understanding on VideoMME v1 (test)
77.2AccuracyGPT4-o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT4-oSize=–, # Frames (Max)=1fps, Budget |M|=–2026.05 | 77.2 | |
| CoRDSSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2.5-VL2026.05 | 65.7 | |
| CoRDSSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2.5-VL2026.05 | 65.3 | |
| Qwen2.5-VL (Original)Size=7B, # Frames (Max)=768, Budget |M|=50K, Backbone=Qwen2.5-VL2026.05 | 64.9 | |
| CoRDSSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2-VL2026.05 | 64.8 | |
| CoRDSSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2-VL2026.05 | 64.3 | |
| Qwen2-VL (Original)Size=7B, # Frames (Max)=768, Budget |M|=50K, Backbone=Qwen2-VL2026.05 | 63.9 | |
| Qwen2-VL + InfiniPot-VSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2-VL2026.05 | 62.8 | |
| LLaVA-Next-Video (Original)Size=7B, # Frames (Max)=128, Budget |M|=25K, Backbone=LLaVA-Next-Video2026.05 | 62.8 | |
| CoRDSSize=7B, # Frames (Max)=128, Budget |M|=6K, Backbone=LLaVA-Next-Video2026.05 | 62.5 | |
| Qwen2-VL + StreamMemSize=7B, # Frames (Max)=4.0/0.5 fps, Budget |M|=6K, Backbone=Qwen2-VL2026.05 | 62.1 | |
| CoRDSSize=7B, # Frames (Max)=128, Budget |M|=3K, Backbone=LLaVA-Next-Video2026.05 | 62 | |
| CoRDSSize=3B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2.5-VL2026.05 | 61.7 | |
| Qwen2-VL + InfiniPot-VSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2-VL2026.05 | 61.2 | |
| LLaVA-Next-Video + InfiniPot-VSize=7B, # Frames (Max)=128, Budget |M|=6K, Backbone=LLaVA-Next-Video2026.05 | 61.1 | |
| CoRDSSize=3B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2.5-VL2026.05 | 61.1 | |
| GPT4-VSize=–, # Frames (Max)=1fps, Budget |M|=–2026.05 | 60.7 | |
| LongVUSize=7B, # Frames (Max)=1fps, Budget |M|=8K2026.05 | 60.6 | |
| Qwen2.5-VL (Original)Size=3B, # Frames (Max)=768, Budget |M|=50K, Backbone=Qwen2.5-VL2026.05 | 60.3 | |
| Qwen2-VL + STCSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2-VL2026.05 | 60.1 | |
| Qwen2-VL + TTCSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2-VL2026.05 | 59.8 | |
| Qwen2.5-VL + StreamMemSize=3B, # Frames (Max)=4.0/0.5 fps, Budget |M|=6K, Backbone=Qwen2.5-VL2026.05 | 59.5 | |
| Qwen2.5-VL + InfiniPot-VSize=3B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2.5-VL2026.05 | 59.3 | |
| LLaVA-OVSize=7B, # Frames (Max)=32, Budget |M|=8K2026.05 | 58.2 | |
| Qwen2-VL + STCSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2-VL2026.05 | 56.1 | |
| Qwen2-VL + TTCSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2-VL2026.05 | 55.3 | |
| LongVUSize=3B, # Frames (Max)=1fps, Budget |M|=8K2026.05 | 51.5 |