ResearchTasksInteractive Tool-Use Agent PerformanceInteractive Tool-Use Agent PerformanceFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedtau2-BenchGPT-5-think81.1Retail Performance Score102Mar 4, 2026VitaBenchQwen3-SE-32B65Delivery Score44May 19, 2026