ResearchBenchmarksOffline Reinforcement Learning on Minari SimpleFollow69.1Humanoid ScoreO2O61.8263.7165.667.49May 7, 2026Evaluation ResultsMethodMethodLinksHumanoid ScoreAnt ScoreHalfCheetah ScoreHopper ScoreWalker2d ScoreAverage Score (Simple)O2ONumber of seeds=10Number of seeds=102026.0569.114.813.185.14.437.3RLPDNumber of seeds=10Number of seeds=102026.0565.38854.573.964.669.3SACfDNumber of seeds=10Number of seeds=102026.056484.255.562.65564.3SPEQNumber of seeds=10Number of seeds=102026.0562.384.45464.25463.8SOPENumber of seeds=10Number of seeds=102026.0562.197.760.765.163.569.8