Agentic Tool-use on tau2-bench Retail
82Pass@1
CODEDELEGATOR
Updated 4d ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2026.01 | 82 | 71.2 | 63.4 | 57 | - | |
2026.01 | 80.9 | 70.3 | 64.5 | 60.5 | - | |
2026.01 | 79.6 | 69.9 | 63.4 | 58.8 | - | |
2026.01 | 78.7 | 68.9 | 62.9 | 58.8 | - | |
2026.01 | 75 | 65.3 | 59.3 | 57.1 | - | |
2026.01 | 70.2 | 59 | 50 | 47 | - | |
2026.01 | - | - | - | - | 73.2 | |
2026.01 | - | - | - | - | 61.4 | |
| - | - | - | - | 74.6 | ||
2026.01 | - | - | - | - | 80.9 |