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

Agentic Task Accuracy on ALFWorld

79.1Task Accuracy

Hard final

56.63662.46868.374.132Jun 18, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
79.1
2026.06
71.6
2026.06
71.6
2026.06
64.9
2026.06
57.5