ResearchBenchmarksMathematical Reasoning on AIME25 (pass@8)Follow63.33Pass@8 ScoreOPSD59.866860.765961.66562.5641May 9, 2026Evaluation ResultsMethodMethodLinksPass@8 ScoreOPSD2026.0563.33OPHSDharness=plan-solveharness=plan-solve2026.0563.33OPHSDharness=noneharness=none2026.0561.67GRPO2026.0560CRISP2026.0560