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

Tool Use on API-Bank (test)

92.6Accuracy

Claude 3.5 Sonnet

36.54451.09765.6580.203Jul 31, 2024Nov 4, 2024Feb 9, 2025May 16, 2025Aug 21, 2025Nov 25, 2025Mar 2, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2024.07
92.6
2024.07
92.3
2024.07
91.3
2024.07
90
2024.07
89
2024.07
82.6
2024.07
73.1
2026.03
71.8
67.1
66.7
2026.03
63.5
2024.07
60.9
2024.07
56.5
2024.07
55.8
2026.03
52.4
2026.03
38.7