ResearchTasksNext Step GenerationNext Step GenerationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedPeanoBench 21 incorrect proofsLeanTutor3.7Accuracy4Mar 5, 2026Human Evaluation SubsetCoP (Qwen2.5-VL-7B)65Win Rate2May 15, 2026