Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Understanding on VStar
Loading...
85.86
Accuracy
Qwen2.5-VL
25.9664
41.5157
57.065
72.6143
Oct 30, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Qwen2.5-VL
Size=72B
2025.10
85.86
InternVL3.5
Size=38B
2025.10
76.96
Qwen2.5-VL
Size=7B
2025.10
76.44
GPT-5
Size=-
2025.10
71.73
Gemini 2.5 Flash
Size=-
2025.10
70.68
InternVL3.5
Size=8B
2025.10
68.59
ThinkMorph
Size=7B, think mode=true
2025.10
67.02
GPT-4o
Size=-
2025.10
61.78
Bagel
Size=7B, think mode=true
2025.10
55.49
Janus-pro
Size=7B
2025.10
38.22
Chameleon
Size=7B
2025.10
28.27
Feedback
Search any
task
Search any
task