Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Video Multimodal Understanding on MMVU
Loading...
68.6
Accuracy
SDRL
43.848
50.274
56.7
63.126
Mar 26, 2026
Accuracy
Updated 20d ago
Evaluation Results
Method
Method
Links
Accuracy
SDRL
Training=RL
2026.03
68.6
VideoRFT
Training=SFT+ RL, Inpu...
2026.03
67.3
TW-GRPO
Training=RL
2026.03
65.8
Qwen2.5-VL-7B
Training=None
2026.03
65.4
SDRL
Training=RL, Training...
2026.03
64.8
Video-R1
Training=SFT+ RL
2026.03
64.2
VideoChat-R1
Training=RL
2026.03
64.2
Video-R1
Training=RL
2026.03
63.8
VideoRFT
Training=RL, Input fra...
2026.03
63.5
Qwen2.5-VL-7B
Training=None, Chain-o...
2026.03
63.2
VideoRFT
Training=SFT, Input fr...
2026.03
60.5
Qwen2.5-VL-7B
Training=None, Chain-o...
2026.03
59.2
Video-R1
Training=SFT
2026.03
51.3
LLaVA-OneVision-7B
Training=None
2026.03
49.2
VideoLLaMA2
Training=None
2026.03
44.8
Feedback
Search any
task
Search any
task