Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

GUI-Odyssey

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mobile GUI AutomationGUI-Odyssey
Success Rate (SR)87
50
Mobile Agent NavigationGUI Odyssey 1.0 (test)
Step Success Rate (SR)81.7
15
GUI GenerationGUI Odyssey OOD
Sad93.45
14
Long-term PlanningGUI-Odyssey
Type Success Rate66.26
14
Mobile Use NavigationGUI Odyssey static offline benchmark
Success Rate83.4
9
GUI Action GroundingGUI-Odyssey (test)
Type Accuracy84.47
8
Mobile Agent EvaluationGUI-Odyssey (test)
Grounding87.02
8
GUI NavigationGUI Odyssey 15 (Overall)
Success Rate73.97
7
GUI Agent Navigation and ActionGUI-Odyssey
Type Accuracy88.54
7
GUI AgentGUI-Odyssey
Type Accuracy90.74
7
Android GUI NavigationGUI Odyssey
Success Rate73.97
5
GUI NavigationGUI-Odyssey High
Action Matching Score (AMS)75.8
4
Showing 12 of 12 rows