ResearchDatasetsLifelong Agent BenchFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsOS TaskLifelong Agent Bench OS Task78.8Success Rate (Last Epoch)31DB TaskLifelong Agent Bench DB Task86Success Rate20DB TaskLifelong Agent Bench (val)94.2Success Rate6OS TaskLifelong Agent Bench (val)74.6Success Rate6