ResearchTaskstask generationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast Updatedred-wine benchmarkgpt-5.461.84Utility9May 11, 2026AZRPROPEL19.95Utility6Jun 18, 2026MathPROPEL41.93Utility Score4Jun 18, 2026SWE OOD (held-out repositories)PROPEL (OOD)19.59Utility2Jun 18, 2026SWE RL-training repositoriesPROPEL26.64Utility2Jun 18, 2026