Offline Reinforcement Learning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
113Average Normalized Reward
19
May 26, 2026
106Average Normalized Reward
19
May 26, 2026
114.7Normalized Score
19
Jun 15, 2026
87Average Success Rate
19
Mar 4, 2026
69Average Success Rate
19
Mar 4, 2026
112.8Normalized Score
19
Feb 26, 2026
82.6Normalized Average Return
19
Jun 4, 2026
105.9Normalized Return
19
Jun 4, 2026
77.6HalfCheetah-Medium Score
19
Jun 9, 2026
76.533Average Return
18
Mar 4, 2026
42.405Average Return
18
Mar 4, 2026
56.213Average Return
18
Mar 4, 2026
47.303Average Return
18
Mar 4, 2026
77.279Average Return
18
Mar 4, 2026
93.704Average Return
18
Mar 4, 2026
87.491Average Return
18
Mar 4, 2026
83.692Average Return
18
Mar 4, 2026
109.803Average Return
18
Mar 4, 2026
93.704Average Return
18
Mar 4, 2026
78.325Average Return
18
Mar 4, 2026
118.564Average Return
18
Mar 4, 2026
87.491Average Return
18
Mar 4, 2026
84.582Average Return
18
Mar 4, 2026
108.1Normalized Score
18
Mar 18, 2026
13.8Normalized Return
18
Feb 26, 2026