ResearchTasksMulti-domain Language Understanding and ReasoningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedOpenCompass MATH, GSM8K, HumanEval, MMLU, CEVAL, CMMLU, BoolQ, CSQA (test)Qwen1.5-MoE4Average K12May 15, 2026
OpenCompass MATH, GSM8K, HumanEval, MMLU, CEVAL, CMMLU, BoolQ, CSQA (test)Qwen1.5-MoE4Average K12May 15, 2026