H1balance_hard on Humanoid-Bench benchmark leaderboard · SOTA2 Research

H1balance_hard on Humanoid-Bench

73Total Average ReturnDual-Flow RL

Evaluation Results

Method	Links
Dual-Flow RL2026.06		73
BRO2026.06		59
QVPO2026.06		57
TD32026.06		56
FlowRL2026.06		55
SAC2026.06		54
QSM2026.06		50
D4PG2026.06		49
MPO2026.06		49
BRO-Fast2026.06		47
DDPG2026.06		46
A2C2026.06		46
TRPO2026.06		39