Offline Reinforcement Learning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
64.4Normalized Reward
24
Feb 26, 2026
108Normalized Reward
24
Feb 26, 2026
74.5Normalized Reward
24
Feb 26, 2026
72.1Avg Normalized Score
24
Feb 26, 2026
121.6Normalized Avg Return
24
Feb 26, 2026
96.7Average Return
24
Feb 26, 2026
91Normalized Score
23
Mar 4, 2026
35.4Normalized Score
23
May 12, 2026
100Normalized Score
23
Feb 26, 2026
95.5Score
22
May 19, 2026
101.3Normalized Return
22
May 19, 2026
111.9Normalized Score
22
Feb 26, 2026
65.9HalfCheetah-Medium
22
Jun 2, 2026
59Normalized Score
21
Jun 2, 2026
84Normalized Score
21
Jun 2, 2026
62.1Mean Normalized Score
21
May 7, 2026
3,718Clean Score
21
Feb 26, 2026
102.29Hopper Score (m)
21
Mar 25, 2026
140.6Normalized Performance Score
20
Jun 15, 2026
137.7Adroit Mean Score
20
Jun 24, 2026
54.9HalfCheetah (Medium)
20
May 15, 2026
94.1Normalized Return
20
Feb 26, 2026
54.1Normalized Return
20
Feb 26, 2026
65.6Normalized Return
20
Feb 26, 2026
118Average Normalized Reward
19
May 26, 2026