Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Agent Task Completion on τ2-BENCH (test)

0.921Average Task Reward

H-EPM

0.127480.333490.53950.74551Dec 8, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2025.12
0.921-
2025.12
0.8420
2025.12
0.52632.8
2025.12
0.51243
2025.12
0.45214.1
2025.12
0.44111.4
2025.12
0.43922.6
2025.12
0.42518.7
2025.12
0.4226.6
2025.12
0.41315.4
2025.12
0.3960
2025.12
0.36226.1
2025.12
0.3580
2025.12
0.33316
2025.12
0.3097.7
2025.12
0.30953.7
2025.12
0.2983.8
2025.12
0.2870
2025.12
0.25627.4
2025.12
0.24622.4
2025.12
0.23246.8
2025.12
0.2010
2025.12
0.19322.2
2025.12
0.189-6
2025.12
0.18416.5
2025.12
0.17410.1
2025.12
0.1580