ResearchTasksLLM Agent Task CompletionLLM Agent Task CompletionFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAgentDojo No AttackIPIGuard73.91Benign Utility22Mar 12, 2026τ-Bench Average across Retail and AirlineDeepSeek-V3.118.1Success Rate (%)8Jun 4, 2026