List your company

Customer Support Interaction on τc-Bench Retail Ours (test)

61Pass@1GPT-5.2
25.6434.824453.18May 27, 2026

Evaluation results

MethodLinks
GPT-5.2User Simulator=GPT-5.22026.0561
GPT-5.2User Simulator=Gemini-3-flash2026.0559
Gemini-3-flashUser Simulator=GPT-5.22026.0555
claude-sonnet-4.6User Simulator=Gemini-3-flash2026.0554
deepseek-3.1User Simulator=GPT-5.22026.0554
Gemini-2.5-flashUser Simulator=GPT-5.22026.0550
deepseek-3.1User Simulator=Gemini-3-flash2026.0547
Gemini-3-flashUser Simulator=Gemini-3-flash2026.0544
Qwen-32BUser Simulator=GPT-5.22026.0539
Gemini-2.5-flashUser Simulator=Gemini-3-flash2026.0536
Qwen-32BUser Simulator=Gemini-3-flash2026.0527