List your company

Customer Support Interaction on τ-Bench Airline Verified (test)

82Pass@1Gemini-3-flash
34.1646.585971.42May 27, 2026

Evaluation results

MethodLinks
Gemini-3-flashUser Simulator=GPT-5.22026.058268
Gemini-3-flashUser Simulator=Gemini-3-flash2026.057256
claude-sonnet-4.6User Simulator=Gemini-3-flash2026.0572—
Gemini-2.5-flashUser Simulator=GPT-5.22026.056640
Gemini-2.5-flashUser Simulator=Gemini-3-flash2026.055840
GPT-5.2User Simulator=GPT-5.22026.055836
GPT-5.2User Simulator=Gemini-3-flash2026.055734
Qwen-32BUser Simulator=Gemini-3-flash2026.055026
deepseek-3.1User Simulator=Gemini-3-flash2026.054930
deepseek-3.1User Simulator=GPT-5.22026.054824
Qwen-32BUser Simulator=GPT-5.22026.053622