Offline Reinforcement Learning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
98.84Avg Normalized Score
7
Feb 26, 2026
80.9Average Normalized Score
7
Feb 26, 2026
90.8Return (HalfCheetah, Medium-Expert)
7
Feb 26, 2026
0.74Average True Return
7
Feb 26, 2026
75Avg True Return
7
Feb 26, 2026
1,922.07Average True Return
7
Feb 26, 2026
108.6Normalized Avg Score
7
Feb 26, 2026
112.4Normalized Average Score
7
Feb 26, 2026
107.6Normalized Score
7
Feb 26, 2026
0.35Normalized Return
7
Feb 26, 2026
10.7Normalized Return
7
Feb 26, 2026
3.1Normalized Return
7
Feb 26, 2026
65Average Task Success
6
May 12, 2026
28Average Task Success
6
May 12, 2026
51Average Task Success
6
May 12, 2026
71Normalized Return
6
May 12, 2026
14Success Rate
6
May 12, 2026
2.97Temporal Error
6
Apr 28, 2026
107Returns
6
Apr 9, 2026
102Returns
6
Apr 9, 2026
97Returns
6
Apr 9, 2026
100Returns
6
Apr 9, 2026
100Returns
6
Apr 9, 2026
90Returns
6
Apr 9, 2026
33.35Normalized Score
6
Mar 18, 2026