List your company

Long Video Understanding on MLVU 2024 (test)

54.9AccuracyGPT-4o
29.10835.80442.549.196Aug 28, 2025

Evaluation results

MethodLinks
GPT-4oSize=-, Frames=0.5 fps / 3842025.0854.9
Video-MTRSize=7B, Frames=802025.0850.4
Video-MTRSize=7B, Frames=642025.0849.8
Video-MTRSize=7B, Frames=322025.0848.4
Video-R1Size=7B, Frames=642025.0847.6
VideoLLaMA2Size=8×7B, Frames=82025.0845.6
Video-XLSize=7B, Frames=128 / 2562025.0845.6
Video-R1Size=7B, Frames=322025.0845.4
Qwen2.5-VL-7BSize=7B, Frames=802025.0845.2
Video-CCAMSize=9B, Frames=962025.0842.9
Qwen2.5-VL-7BSize=7B, Frames=642025.0841.8
Qwen2.5-VL-7BSize=7B, Frames=322025.0841.6
LongVASize=7B, Frames=128 / 2562025.0841.1
Video-LLaVASize=7B, Frames=82025.0830.7
VideoChat2Size=7B, Frames=162025.0830.1