ResearchTasksCode DebuggingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedHumanEvalMGDebugger96.3Accuracy42Feb 26, 2026MBPPMGDebugger80.8Accuracy30Feb 26, 2026InfiniteBench code_debug 40k input capQwen3-4B34.26Accuracy19May 27, 2026DebugBenchMOCHA66.6Average Accuracy11May 20, 2026LiveCodeBench February-May 2025 (evaluation window)MGDebugger36.6Accuracy4Feb 26, 2026