ResearchTasksDeep-search QAFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedXbench-DeepSearch (test)Tongyi-DR-30B75Pass@124Feb 26, 2026BrowseComp (test)OpenAI-DeepResearch51.5Pass@124Feb 26, 2026FramesProCeedRL46.42Accuracy16Jul 1, 2026MuSiQueProCeedRL29.52Accuracy8Apr 3, 2026WebwalkerQAProCeedRL23.01Accuracy6Apr 3, 2026