Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Perception and Grounding on V*
Loading...
88.9
Accuracy
DeepEyes
38.148
51.324
64.5
77.676
May 13, 2026
Accuracy
Updated 19d ago
Evaluation Results
Method
Method
Links
Accuracy
DeepEyes
Size=7B
2026.05
88.9
MoCA
Size=7B
2026.05
86.6
Pixel Reasoner
Size=7B
2026.05
84.3
Qwen2.5-VL-Instruct
Size=72B
2026.05
81.2
Llava-OV
Size=7B
2026.05
72.8
Qwen2.5-VL-Instruct
Size=7B
2026.05
71.4
VL-Rethinker
Size=7B
2026.05
68.2
mPLUG-Owl3
Size=7B
2026.05
64.5
R1-VL
Size=7B
2026.05
60.3
GPT-4o-mini
Size=-
2026.05
50.8
GPT-4o
Size=-
2026.05
45
Docopilot
Size=8B
2026.05
40.1
Feedback
Search any
task
Search any
task