Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
GUI Operation on GUIAct Web-Multi (test)
Loading...
84.1
Type EM
MFT
66.316
70.933
75.55
80.167
Feb 14, 2026
Type EM
Cli.Acc
StepSR
Updated 1mo ago
Evaluation Results
Method
Method
Links
Type EM
Cli.Acc
StepSR
MFT
Backbone=Qwen2.5-VL-72B
2026.02
84.1
67.3
73.6
MFT
Backbone=Qwen2.5-VL-7B
2026.02
83.3
64.4
71.9
Qwen-GUI
Backbone=Qwen-VL
2026.02
68.9
52.5
46.8
MiniCPM-GUI
Backbone=MiniCPM-V
2026.02
67
45.5
47.5
Feedback
Search any
task
Search any
task