Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

OS-Kairos

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mobile Task AutomationOS-Kairos
Scroll Rate66.67
10
Mobile-use agent task completion and intent alignmentOS-Kairos user-specific Modified
Success Rate (SR)69.74
8
Mobile Task ExecutionOS-Kairos real-world Android devices (test)
RE Rate93.09
5
GUI NavigationOS-Kairos 1.0 (test)
Scroll Success Rate100
4
Showing 4 of 4 rows