ResearchTasksVision Agent CapacityVision Agent CapacityFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedHLE-VLSeed2.0 Pro39.2Score6Jul 2, 2026MM-BrowseCompSeed2.0 Pro48.8Score6Jul 2, 2026MineDojo VerifiedSeed2.0 Pro49Score6Jul 2, 2026