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

Tool-use Task Completion on ACEBench agent subset Multi-Turn

84.4Success Rate

ReAct + EvoSOP

45.19255.37165.5575.729Jul 8, 2026
Updated 17d ago

Evaluation Results

MethodLinks
2026.07
84.4
2026.07
83.3
2026.07
79.4
2026.07
79.4
2026.07
78.3
2026.07
77.9
2026.07
74.4
2026.07
72.2
2026.07
71.1
2026.07
69.4
2026.07
66.1
2026.07
65.6
2026.07
57.2
2026.07
46.7