ResearchBenchmarksStand on Humanoid-benchFollow815.9Maximum ReturnIRM-2.684209.833422.35634.867Mar 9, 2026Evaluation ResultsMethodMethodLinksMaximum ReturnConvergence StepsIRM2026.03815.90.6TD-MPC22026.03749.81.8DreamerV32026.03699.35.5SAC2026.0393.64.9PPO2026.0328.85.3