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

Agent Task on AppWorld Challenge (test)

66Task Goal Completion (TGC)

ReAct + ACE

29.39238.89648.457.904Oct 6, 2025
Updated 2mo ago

Evaluation Results

MethodLinks
2025.10
6648.9
2025.10
57.339.6
2025.10
54.435.2
2025.10
52.330.8
2025.10
4627.3
2025.10
4630.2
2025.10
43.220.1
2025.10
41.521.6
2025.10
40.320.9
2025.10
40.120.9
2025.10
39.618.7
2025.10
34.515.1
2025.10
30.818.2