Tau3-Bench Airline: leaderboard

τ³-bench airline domain: multimodal, knowledge-aware agentic customer service tasks. New domain in τ³-bench. From Sierra AI.

Metric: Pass@1 (%). Source: taubench.com. Status: saturation imminent. 8 models tracked.

Top models

#ModelScore
1Claude Opus 4.584
2GPT-5.283
3GLM-582.5
4Gemini 3 Flash82.5
5Qwen 3.5 397B A17B81.5
6Gemini 3 Pro80.5
7Claude Sonnet 4.572

Interactive version: theaggregate.ai/benchmark?slug=tau3-bench-airline · How It Works · Data refreshed daily, snapshot 2026-09-05.