Offline Reinforcement Learning on AntMaze Ultra-Diverse benchmark leaderboard · SOTA2 Research

Offline Reinforcement Learning on AntMaze Ultra-Diverse

5,460Avg Normalized ScoreGCPC

Evaluation Results

Method	Links
GCPC2023.07		5,460
GCIQL2023.07		2,840
RvS-G2023.07		2,640
TAP2023.07		2,600
DWSL2023.07		2,500
IQL2023.07		1,780
CQL2023.07		960
LEQ2025.04		55.8
VIPO-LEQ2025.04		26.5
IQL-TD-MPC2025.04		3.6
BC2023.07		0
DT2023.07		0
WGCSL2023.07		0
MOBILE2025.04		0