ResearchBenchmarksHumanoid-run on Humanoid-BenchFollow433Total Average ReturnDual-Flow RL-16.28100.36217333.64Jun 29, 2026Evaluation ResultsMethodMethodLinksTotal Average ReturnDual-Flow RL2026.06433BRO2026.06376BRO-Fast2026.06364FlowRL2026.06329TD32026.06329SAC2026.06204QSM2026.0698D4PG2026.0640DDPG2026.062MPO2026.062TRPO2026.062QVPO2026.061A2C2026.061