Customer Support Interaction on τc-Bench Airline Ours (test)
64Pass@1claude-sonnet-4.6
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| claude-sonnet-4.6User Simulator=Gemini-3-flash2026.05 | 64 | — | |
| Gemini-3-flashUser Simulator=GPT-5.22026.05 | 56 | 28 | |
| GPT-5.2User Simulator=Gemini-3-flash2026.05 | 49 | 26 | |
| deepseek-3.1User Simulator=Gemini-3-flash2026.05 | 41 | 16 | |
| Gemini-2.5-flashUser Simulator=GPT-5.22026.05 | 36 | 8 | |
| deepseek-3.1User Simulator=GPT-5.22026.05 | 36 | 18 | |
| Gemini-3-flashUser Simulator=Gemini-3-flash2026.05 | 34 | 22 | |
| GPT-5.2User Simulator=GPT-5.22026.05 | 34 | 20 | |
| Gemini-2.5-flashUser Simulator=Gemini-3-flash2026.05 | 21 | 10 | |
| Qwen-32BUser Simulator=Gemini-3-flash2026.05 | 13 | 6 | |
| Qwen-32BUser Simulator=GPT-5.22026.05 | 10 | 6 |