Offline Reinforcement Learning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
89.2Medium Diverse Success Rate
27
May 27, 2026
91.1Average Normalized Score
27
Jun 2, 2026
85Score
27
May 29, 2026
4,452Average Score
27
Mar 18, 2026
56Success Rate
26
May 12, 2026
94.7Normalized Performance
26
Jun 11, 2026
89.22Score
26
Mar 4, 2026
111.587Score
26
Mar 4, 2026
98.988Score
26
Mar 4, 2026
101.244Score
26
Mar 4, 2026
113.069Score
26
Mar 4, 2026
71.67Performance Score
26
Mar 4, 2026
80.693Score
26
Mar 4, 2026
71.9HalfCheetah (m)
26
Jun 11, 2026
107.6Normalized Score
26
May 29, 2026
101.6Normalized Return
26
Feb 26, 2026
92.5Normalized Return
26
Mar 19, 2026
86.7Normalized Score
26
Feb 26, 2026
24Success Rate
25
May 12, 2026
98.1Normalized Score
25
Jun 2, 2026
91.6Normalized Return
25
Feb 26, 2026
96.6Normalized Performance
24
Jun 4, 2026
71.48Target Domain Score
24
Feb 26, 2026
64.31Normalized Target-Domain Score
24
Feb 26, 2026
100Normalized Score
24
May 29, 2026