Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
GUI agent task on OSWORLD clean (Trained (86))
Loading...
87.2
pass@8 Success Rate
UI-TARS-1.5-7B + ENVS
72.744
76.497
80.25
84.003
Jun 22, 2026
pass@8 Success Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
pass@8 Success Rate
UI-TARS-1.5-7B + ENVS
Model=UI-TARS-1.5-7B +...
2026.06
87.2
UI-TARS-1.5-7B + ARPO
Model=UI-TARS-1.5-7B +...
2026.06
81.4
UI-TARS-1.5-7B
Model=UI-TARS-1.5-7B,...
2026.06
73.3
Feedback
Search any
task
Search any
task