ResearchTasksUtilityFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedG2U UtilityHuman1Mean Utility30May 15, 2026MMLUGPT-OSS-20B83.1Accuracy23May 28, 2026OPI cleanGemma-4-E4B-it90.3Utility Score15May 28, 2026POPENeural Gate85.56Score14Mar 16, 2026MMEDINM0.729Score14Mar 16, 2026ScienceQADINM61.33Score14Mar 16, 2026