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

GUI Automation on OSWorld

86.8Overall Success Rate

Agentic Programming (Ours)

24.71240.83156.9573.069Jun 14, 2026Jun 17, 2026Jun 21, 2026Jun 24, 2026Jun 28, 2026Jul 1, 2026Jul 5, 2026
Updated 19d ago

Evaluation Results

MethodLinks
2026.06
86.893.580100
2026.06
80.478.362.995.8
2026.06
78.371.765.783.3
2026.06
76.366.661.191.7
2026.06
74.560.464.387.5
2026.06
73.176.75579.2
2026.06
72.66363.979.2
2026.06
72.178.560.291.7
2026.06
69.969.558.779.2
2026.06
67.169.55370.8
2026.06
6661.757.475
2026.07
38.2---
2026.07
38.1---
2026.07
36.8---
2026.07
36.5---
2026.07
35---
2026.07
34.9---
2026.07
34.8---
2026.07
34.1---
2026.07
33.9---
2026.07
33.9---
2026.07
32.6---
2026.07
31.9---
2026.07
31.6---
2026.07
31.3---
2026.07
28.2---
2026.07
27.4---
2026.07
27.1---