ResearchBenchmarksFormal Theorem Proving on NuminaMath LEAN (unsolved)Follow26AccuracyAx-Prover-1.045.981320.02Oct 14, 2025Evaluation ResultsMethodMethodLinksAccuracyAx-Prover2025.1026DS-Prover2025.1018Sonnet2025.101Kiminaevaluation_protocol=pa...evaluation_protocol=pass@682025.100