GUI Agent Task
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
80Success Rate
200
Jul 1, 2026
37.5Precision@1
19
Jun 19, 2026
34.2P@1
19
Jun 19, 2026
35.7Precision@1
19
Jun 19, 2026
41.7P@1
19
Jun 19, 2026
88.6Success Rate
9
May 13, 2026
50.8Completion Rate
4
Feb 26, 2026
29Pass@8 Success Rate
3
Jun 23, 2026
5.1Pass@8 Success Rate
3
Jun 23, 2026
88.4Pass@8 Success Rate
3
Jun 23, 2026
30.3Pass@8 Success Rate
3
Jun 23, 2026
7.5Pass@8 Success Rate
3
Jun 23, 2026
87.2pass@8 Success Rate
3
Jun 23, 2026
0.95Success Rate
2
Feb 26, 2026