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

Agentic Task Completion on AppWorld Challenge (test)

49.88Task Goal Completion (TGC)

Claude-Sonnet-4.5

-1.246412.026825.338.5732Apr 17, 2026
Updated 3mo ago

Evaluation Results

MethodLinks
2026.04
49.8832.37
2026.04
40.0520.14
2026.04
31.6516.55
2026.04
22.789.35
2026.04
19.95.76
2026.04
17.036.47
2026.04
16.314.32
2026.04
14.69.3
2026.04
9.830.72
2026.04
8.392.16
2026.04
7.912.16
2026.04
5.520
2026.04
0.720