ResearchTasksAggregated Logical ReasoningAggregated Logical ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedOverall MeanDeepseek-V3.2-R76.2Accuracy7Feb 26, 2026Overall UnsolvableGPT-5.1-Low0.945Accuracy7Feb 26, 2026Overall SolvableDeepseek-V3.2-R67.3Accuracy7Feb 26, 2026