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

Tool-use on Tau-Bench (Pass@1 Domain Scores and Word Count)

67.42Average Pass@1

GLM-4.5

3.449620.057336.66553.2727Aug 12, 2025Sep 16, 2025Oct 22, 2025Nov 27, 2025Jan 2, 2026Feb 7, 2026Mar 15, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.03
67.4255.3579.540,300
2026.03
64.9154.4575.3742,000
2026.03
62.1244.0580.2-
2026.03
59.3548.0569.54-
2026.03
59.2550.768-
2026.03
53.754265.5127,000
2026.03
50.5237.3563.745,000
2025.08
50.22---
2026.03
50.0136.9563.0745,000
2025.08
47.09---
2026.03
4327.9558.0439,000
2025.08
40.04---
2026.03
37.9317.1558.7230,000
2026.03
37.9117.458.4129,700
2025.08
37.43---
2025.08
34.26---
2025.08
32.52---
2025.08
31.7---
2025.08
31---
2025.08
28.91---
2025.08
27.39---
2025.08
27.09---
2025.08
26.83---
2025.08
25.43---
2025.08
24.26---
2025.08
23.35---
2025.08
21.96---
2025.08
21.91---
2025.08
18.87---
2025.08
17.61---
2025.08
16.74---
2025.08
16.43---
2025.08
15.13---
2025.08
13.74---
2025.08
11.91---
2025.08
10.13---
2025.08
6.91---
2025.08
5.91---