Offline Reinforcement Learning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
40Normalized Score
11
Apr 27, 2026
69Normalized Score
11
Apr 27, 2026
11Normalized Score
11
Apr 27, 2026
68Normalized Score
11
Apr 27, 2026
90Normalized Score
11
Apr 27, 2026
113.3Walker2d (Medium Expert) Score
11
Mar 19, 2026
92.7Visual Cube Single Task 1 Success Rate
11
Mar 17, 2026
51.8Score
11
Feb 26, 2026
74.1HalfCheetah-Medium Return
11
Feb 26, 2026
70Normalized Score
11
Feb 26, 2026
79.5Normalized Score
11
Feb 26, 2026
84Normalized Score
11
Feb 26, 2026
92.7Normalized Score
11
Feb 26, 2026
111.8Normalized Return
11
Feb 26, 2026
75.1Normalized Return
10
May 29, 2026
79.2Normalized Return
10
May 29, 2026
83.5Normalized Return
10
May 29, 2026
87.1Normalized Return
10
May 29, 2026
67.5Normalized Return
10
May 29, 2026
71.3Normalized Return
10
May 29, 2026
75.8Normalized Return (%)
10
May 29, 2026
84.2Normalized Return
10
May 29, 2026
79.3Normalized Return
10
May 29, 2026
82.4Normalized Return
10
May 29, 2026
87.9Normalized Return
10
May 29, 2026