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

Web Task Completion on MiniWoB++ Held-out Goals (test)

70.5Success Rate

Claude-3.5-Sonnet

27.8638.935061.07Jul 5, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.07
70.5
2025.07
69.7
2025.07
67.2
2025.07
65.9
2025.07
65.7
2025.07
65
2025.07
63.2
2025.07
61
2025.07
57
2025.07
57
2025.07
56.2
2025.07
55.2
2025.07
52.5
2025.07
43.5
2025.07
32.8
2025.07
29.5