Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
GUI Grounding on OSWorld
Loading...
44.1
Success Rate
GPT-5.4
37.444
39.172
40.9
42.628
May 12, 2026
Success Rate
Updated 21d ago
Evaluation Results
Method
Method
Links
Success Rate
GPT-5.4
Planner=GPT-5.4, max a...
2026.05
44.1
Phi-Ground-Any-4B
Planner=GPT-5.4, max a...
2026.05
42.4
MAI-UI-8B
Planner=GPT-5.4, max a...
2026.05
38.2
GUI-Owl-1.5-8B-Instruct
Planner=GPT-5.4, max a...
2026.05
37.7
Feedback
Search any
task
Search any
task
