ResearchTasksComprehensive ExaminationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedAGIEval (test)GPT 4o62.3Accuracy37May 19, 2026GAOKAO (test)Qwen-14B-Chat76.3Accuracy27Feb 26, 2026C-Eval (test)Qwen-14B-Chat71.5Accuracy14Feb 26, 2026CMMLU (test)Qwen-14B-Chat68.1Accuracy14Feb 26, 2026