ResearchTasksMultitask Language UnderstandingMultitask Language UnderstandingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMMLU ProQwen3.5-2B66.5Score2Jun 4, 2026MMMLUHATified59.7Normalized Log Accuracy2Mar 18, 2026PreviousPage 4 of 4Next