List your company
ResearchFormerly WizWand

Long Video Understanding on LongVideoBench 2024 (val)

66.7AccuracyGPT-4o
37.99645.44852.960.352Aug 28, 2025
Updated 5mo ago

Evaluation results

MethodLinks
GPT-4oSize=-, Frames=0.5 fps / 3842025.0866.7
Gemini-1.5-ProSize=-, Frames=0.5 fps2025.0864
Video-MTRSize=7B, Frames=802025.0857.1
LLaVA-OneVisionSize=7B, Frames=322025.0856.3
Qwen2.5-VL-7BSize=7B, Frames=7682025.0856
Video-MTRSize=7B, Frames=642025.0854.8
Video-MTRSize=7B, Frames=322025.0852.3
Video-XLSize=7B, Frames=128 / 2562025.0850.7
Qwen2.5-VL-7BSize=7B, Frames=802025.0848.4
LongVASize=7B, Frames=128 / 2562025.0847.8
Qwen2.5-VL-7BSize=7B, Frames=642025.0847
Qwen2.5-VL-7BSize=7B, Frames=322025.0845.8
Video-CCAMSize=9B, Frames=962025.0843.1
VideoChat2Size=7B, Frames=162025.0839.3
Video-LLaVASize=7B, Frames=82025.0839.1