ResearchTasksCode UnderstandingCode UnderstandingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedCRUX (test)ReM-MoA*69.3Pass@130Jun 24, 2026CodeMMLUW/O (FP16)53.52Exact Match (EM)28May 19, 2026RepoQA and CodeQueries zero-shotText65.6Accuracy12May 11, 2026MLE-Bench AI4CodeScientistOne83.56Score2May 27, 2026