List your company

Offline Video Understanding on MLVU v1 (test)

71.5AccuracyCoRDS
54.13258.64163.1567.659May 14, 2026

Evaluation results

MethodLinks
CoRDSSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2.5-VL2026.0571.5
CoRDSSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2-VL2026.0570.1
CoRDSSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2.5-VL2026.0570
CoRDSSize=3B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2.5-VL2026.0569.6
LLaVA-Next-Video (Original)Size=7B, # Frames (Max)=128, Budget |M|=25K, Backbone=LLaVA-Next-Video2026.0568.7
Qwen2.5-VL (Original)Size=7B, # Frames (Max)=768, Budget |M|=50K, Backbone=Qwen2.5-VL2026.0568.5
CoRDSSize=3B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2.5-VL2026.0568.3
CoRDSSize=7B, # Frames (Max)=128, Budget |M|=6K, Backbone=LLaVA-Next-Video2026.0568.1
CoRDSSize=7B, # Frames (Max)=128, Budget |M|=3K, Backbone=LLaVA-Next-Video2026.0567.8
CoRDSSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2-VL2026.0567.4
GPT4-oSize=–, # Frames (Max)=1fps, Budget |M|=–2026.0566.2
Qwen2-VL + StreamMemSize=7B, # Frames (Max)=4.0/0.5 fps, Budget |M|=6K, Backbone=Qwen2-VL2026.0565.9
Qwen2-VL (Original)Size=7B, # Frames (Max)=768, Budget |M|=50K, Backbone=Qwen2-VL2026.0565.8
Qwen2-VL + InfiniPot-VSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2-VL2026.0565.8
LongVUSize=7B, # Frames (Max)=1fps, Budget |M|=8K2026.0565.4
LLaVA-Next-Video + InfiniPot-VSize=7B, # Frames (Max)=128, Budget |M|=6K, Backbone=LLaVA-Next-Video2026.0565.2
LLaVA-OVSize=7B, # Frames (Max)=32, Budget |M|=8K2026.0564.7
Qwen2.5-VL (Original)Size=3B, # Frames (Max)=768, Budget |M|=50K, Backbone=Qwen2.5-VL2026.0563.3
Qwen2-VL + InfiniPot-VSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2-VL2026.0563.1
Qwen2.5-VL + StreamMemSize=3B, # Frames (Max)=4.0/0.5 fps, Budget |M|=6K, Backbone=Qwen2.5-VL2026.0562.3
Qwen2.5-VL + InfiniPot-VSize=3B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2.5-VL2026.0562.1
Qwen2-VL + TTCSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2-VL2026.0558.4
Qwen2-VL + STCSize=7B, # Frames (Max)=768, Budget |M|=6K, Backbone=Qwen2-VL2026.0557.9
LongVUSize=3B, # Frames (Max)=1fps, Budget |M|=8K2026.0555.9
Qwen2-VL + STCSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2-VL2026.0555
Qwen2-VL + TTCSize=7B, # Frames (Max)=768, Budget |M|=3K, Backbone=Qwen2-VL2026.0554.8