Windows UI Navigation on WindowsAgentArena (WAA)
74.5Success Rate
Human Performance
Updated 5d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.01 | 74.5 | - | |
2026.01 | 57.5 | 30 | |
2026.01 | 56.6 | 100 | |
2026.01 | 50.3 | 30 | |
2026.01 | 49 | 50 | |
2026.05 | 39.1 | - | |
2026.01 | 37.4 | 50 | |
2026.05 | 35.9 | - | |
2026.05 | 32.9 | - | |
2026.01 | 30.5 | 30 | |
2026.05 | 30.2 | - | |
2026.01 | 29.8 | 15 | |
2026.05 | 28.2 | - | |
2026.05 | 24.9 | - | |
2026.05 | 24.2 | - | |
2026.05 | 21.4 | - | |
2026.01 | 21.1 | 15 | |
2026.05 | 20.7 | - | |
2026.01 | 20.1 | 40 | |
2026.01 | 19.5 | 15 | |
2026.01 | 18.2 | 15 | |
2026.01 | 18.1 | 15 | |
2026.05 | 18 | - | |
2026.05 | 17.9 | - | |
2026.05 | 15.9 | - | |
2026.01 | 15.7 | 50 | |
2026.05 | 13.5 | - | |
2026.05 | 11.8 | - | |
2026.05 | 11.1 | - | |
2026.05 | 11.1 | - | |
2026.05 | 9.7 | - | |
2026.05 | 7.1 | - | |
2026.05 | 6.4 | - |