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

Agentic Tool Use on Tau2-Bench (Domain Scores)

90.4Retail Score

Seed 2.0

-3.61620.79245.269.608Apr 20, 2026Apr 27, 2026May 4, 2026May 11, 2026May 18, 2026May 25, 2026Jun 1, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.04
90.494.264.483
2026.04
86.29870.184.7
2026.04
85.39872.785.4
2026.04
81.695.862.580.2
2026.05
81.196.762-
2026.05
77.1966.6762-
2026.04
74.556.15265.4
2026.05
73.965.851.2-
2026.04
72.850.94061.8
2026.04
71.95845.658.5
2026.05
70.446.546-
2026.05
70.458.252-
2026.05
68.727.244-
2026.04
67.849.24855
2026.05
67.845.646-
2026.05
66.138.540-
2026.04
63.617.831.539
2026.04
59.527.24844.9
2026.04
55.314.92732.4
2026.04
50.927.237.538.5
2026.04
49.632.731.537.9
2026.06
47.48.840-
2026.06
46.515.824-
2026.06
46.515.824-
2026.06
46.519.328-
2026.06
43.917.528-
2026.06
43.917.540-
2026.04
41.238.523.534.4
2026.06
40.421.138-
2026.06
39.51420-
2026.06
38.6716-
2026.06
38.6716-
2026.06
37.77.922-
2026.06
36.811.426-
2026.06
36.816.714-
2026.06
36.816.714-
2026.06
35.115.832-
2026.06
34.27.928-
2026.04
341826.526.2
2026.06
33.34.420-
2026.06
32.515.824-
2026.06
32.55.326-
2026.06
32.513.224-
2026.04
32.229.23431.8
2026.06
30.79.732-
2026.06
25.415.820-
2026.06
25.49.618-
2026.06
25.415.820-
2026.06
23.713.224-
2026.06
23.713.224-
2026.06
23.78.818-
2026.04
22.810.52017.7
2026.06
20.27.918-
2026.06
12.35.314-
2026.06
75.38-
2026.06
000-
2026.06
000-
2026.06
000-
2026.06
000-
2026.04
---80.3