ResearchTasksMulti-discipline Knowledge EvaluationMulti-discipline Knowledge EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedCodaSet ID MMLU-PRO (test)Qwen3-235B-A22B84.86Accuracy16May 26, 2026