TauBench V2 Airline
tau2-bench - Airline domain. Agentic tool use benchmark for airline customer service scenarios.
Category
General Agents
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for TauBench V2 Airline
| Model | Score | Date | Verified |
|---|---|---|---|
| Qwen3-Next-80B-A3B-Thinking |
100.00%
|
10.09.2025 | Verified |