ResearchTasksKnowledge-based Agent ReasoningKnowledge-based Agent ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedKARLBench 1.0 (test)Claude 4.6 Opus75.9BrowseComp-Plus Score17Mar 6, 2026