ResearchTasksPost-training Performance EvaluationPost-training Performance EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedLlama 3.2-3B (val)SFT889K -> DAPOdc70.1Max Ppost15Feb 26, 2026