ResearchTasksSearch Agent CapacityFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSeal-0Claude-Sonnet-4.553.4Score8Jul 2, 2026DeepSearchQASeed2.0 Pro77.4Score8Jul 2, 2026FinSearchCompGPT-5.2 High73.8Score8Jul 2, 2026WideSearchGPT-5.2 High76.8Score8Jul 2, 2026HLE VerifiedSeed2.0 Pro73.6Score8Jul 2, 2026