Agentic Reasoning on TauBench V3
85.8Airline Accuracy
Kimi-K2.6
Updated 1mo ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2026.06 | 85.8 | 82.9 | 97.8 | 23.1 | 72.4 | |
2026.06 | 85 | 84.1 | 96.9 | 12.8 | 69.7 | |
2026.06 | 81.5 | 86.4 | 92.9 | 22.6 | 70.9 | |
2026.06 | 80.8 | 88.9 | 96.3 | 25.9 | 73.2 | |
2026.06 | 80.8 | 89.1 | 98.3 | 26.7 | 73.7 | |
2026.06 | 76.5 | 88.5 | 98 | 20.9 | 71 | |
2026.06 | 75.3 | 84.9 | 89.6 | 14.6 | 66.1 |