ResearchTasksMultilingual Knowledge EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedM-MMLUKarnak77.83Overall Accuracy31Mar 18, 2026Global-MMLU LowGRPO52.31Accuracy12Jun 8, 2026Global-MMLU HighGRPO68.35Accuracy12Jun 8, 2026Multilingual ExamQwen2-72B76.6Accuracy5Feb 26, 2026