Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Reasoning on AI2D
Loading...
83.8
Accuracy
DualMindVLM
80.68
81.49
82.3
83.11
Nov 20, 2025
Accuracy
Length
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Length
DualMindVLM
Size=7B, Strategy=RL
2025.11
83.8
104
ThinkLite
Size=7B, Strategy=RL
2025.11
83.6
168
MM-Eureka
Size=7B, Strategy=RL
2025.11
83.5
207
OpenVLThinker
Size=7B, Strategy=SFT+RL
2025.11
83.2
160
VL-Rethinker
Size=7B, Strategy=RL
2025.11
82.4
226
Qwen2.5-VL
Size=7B, Strategy=-
2025.11
80.8
145
Feedback
Search any
task
Search any
task