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

Agent Task Completion on AppWorld (Resampled)

66.1Accuracy (%)

Metis

54.34857.39960.4563.501Jun 23, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
66.178.510.3211.8
2026.06
65.589.511.0317.8
2026.06
57.7204.811.94.3
2026.06
54.8101.713.920