ResearchTasksCode evaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedBigCodeBenchGPT-o1-preview82.02Accuracy23May 28, 2026CodeMMLU software principlesDPO + RM24.9Accuracy12Jul 8, 2026ruCodeEval 3 (test)T-pro 2.02.15Speedup2Feb 26, 2026