ResearchTasksMulti-agent system task solvingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedworkbenchTacoMAS82.4Accuracy21May 12, 2026plancraftTacoMAS88.7Accuracy21May 12, 2026browsecompTacoMAS74.5Accuracy21May 12, 2026financeTacoMAS76.7Accuracy21May 12, 2026