ResearchBenchmarksMath Reasoning on AIME 2024 (pass@1, pass@4)Follow9.27Pass@1XRPO7.10687.66848.238.7916Oct 8, 2025Evaluation ResultsMethodMethodLinksPass@1Pass@4XRPOModel=Llama-3.2-3BModel=Llama-3.2-3B2025.109.2718.7GRPOModel=Llama-3.2-3BModel=Llama-3.2-3B2025.107.1916.17