Multi-turn tool calling on τ2-bench
17.77Overall Score
Toucan
Updated 4d ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.01 | 17.77 | 20 | 22.8 | 10.5 | |
2026.01 | 17.62 | 38 | 6.1 | 8.77 | |
2026.01 | 17.02 | 30 | 6.14 | 14.91 | |
2026.01 | 16.08 | 14 | 17.54 | 16.7 | |
2026.01 | 13.56 | 24 | 7.02 | 9.65 |
Toucan
| Method | Links | ||||
|---|---|---|---|---|---|
2026.01 | 17.77 | 20 | 22.8 | 10.5 | |
2026.01 | 17.62 | 38 | 6.1 | 8.77 | |
2026.01 | 17.02 | 30 | 6.14 | 14.91 | |
2026.01 | 16.08 | 14 | 17.54 | 16.7 | |
2026.01 | 13.56 | 24 | 7.02 | 9.65 |