ResearchDatasetsMMLU, GPQA, MedMCQA, BBEHMMLU, GPQA, MedMCQA, BBEHFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsLLM RoutingMMLU-PRO, SUPERGPQA, MEDMCQA, BBEH (test)91.5MMLU-PRO Top-1 Acc14