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

Tool Use Accuracy on tau2-Bench

30.7Accuracy

Qwen2.5-32B-CodeGym

1.4769.06316.6524.237Sep 22, 2025Oct 8, 2025Oct 25, 2025Nov 11, 2025Nov 27, 2025Dec 14, 2025Dec 31, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.09
30.7
2025.09
30.7
2025.09
26.1
2025.09
25.8
2025.09
24.7
2025.09
22.6
2025.09
20.9
2025.09
19.9
2025.09
15.5
2025.12
15
2025.09
14.9
2025.12
10.9
2025.12
9.7
2025.12
2.6