ResearchTasksLong-form researchFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedDRBGPT-5 + Search50.7Score39May 18, 2026ResearchQAOpenAI Deep Research79.2Score18May 12, 2026ResearchRubricsGemini Deep Research61.5Score16May 12, 2026HealthBenchGPT-5 + Search59.5Overall Score14May 12, 2026