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

Multi-turn tool-use interaction on tau-bench

86.1Retail Success Rate

U-Fold

16.4234.5152.670.69Dec 4, 2025Dec 12, 2025Dec 21, 2025Dec 30, 2025Jan 8, 2026Jan 17, 2026Jan 26, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.01
86.154-
2026.01
85.2562.8
2026.01
76.5529.5
2026.01
74.854-
2026.01
7361.5-
2026.01
72.2602.5
2026.01
70463.4
2026.01
69.6547.3
2026.01
67461.5
2026.01
65.244-
2026.01
65.25013.9
2025.12
5643-
2026.01
55.74615.8
2026.01
53.9406.2
2025.12
5241-
2025.12
5136-
2025.12
4939-
2026.01
46.14612.7
2025.12
4432-
2025.12
4431-
2026.01
43.54211.3
2025.12
4222-
2025.12
3529-
2025.12
3529-
2025.12
3211-
2025.12
3114-
2025.12
2919-
2025.12
2516-
2026.01
24.34227
2025.12
2414-
2026.01
22.6321.1
2026.01
21.734-
2025.12
2113-
2026.01
20303.7
2026.01
19.1263.7