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

Robust GUI Navigation on GUI-RobustEval

55.8Success Rate (Depth 0)

Jedi-7B w/ GPT 5.1

3.07216.76130.4544.139May 28, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.05
55.844.232.229.134.6
2026.05
55.146.136.933.565.4
2026.05
54.944.634.532.163.9
2026.05
49.741.836.533.258.8
2026.05
48.138.73025-
2026.05
45.537.228.625.950.3
2026.05
43.536.630.126.751.9
2026.05
40.730.323.31946.3
2026.05
39.634.227.823.338
2026.05
28.715.68.110.45.9
2026.05
23.416.513.111.233.9
2026.05
5.132.91.3-