ResearchTasksCustomer Support InteractionCustomer Support InteractionFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedτc-Bench Telecom Ours (test)claude-sonnet-4.672Pass@111May 28, 2026τ-Bench Telecom Verified (test)Gemini-3-flash94Pass@111May 28, 2026τc-Bench Retail Ours (test)GPT-5.261Pass@111May 28, 2026τ-Bench Retail Verified (test)GPT-5.292Pass Rate11May 28, 2026τc-Bench Airline Ours (test)claude-sonnet-4.664Pass@111May 28, 2026τ-Bench Airline Verified (test)Gemini-3-flash82Pass@111May 28, 2026