Reinforcement Learning
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
1,047Median Human-Normalized Score
9
Feb 26, 2026
7,127.7Alien Score
9
Feb 26, 2026
47.5Median Score
9
Feb 26, 2026
32.4Score
9
Feb 26, 2026
1,400Score
9
Feb 26, 2026
19,564.9Asterix Score
9
Feb 26, 2026
34.97Mean Performance
8
Jun 30, 2026
43.8Mean Score
8
Jun 30, 2026
6E1
8
Jun 4, 2026
11.87Max Reward (%)
8
Jun 2, 2026
312.5Average Return
8
May 18, 2026
-33.8Average Return
8
May 18, 2026
972.4Average Return
8
May 18, 2026
-42.3Average Return
8
May 18, 2026
-41.6Average Return
8
May 18, 2026
589.7Steps to 75% Return (k)
8
May 18, 2026
1.219IQM Score
8
May 13, 2026
1.03Mean Reward
8
May 12, 2026
1,000Mean Reward
8
May 12, 2026
1.64IQM Human-Normalized Score
8
May 13, 2026
6.3Coefficient of Variation (%)
8
Mar 13, 2026
3.2Coefficient of Variation
8
Mar 13, 2026
10.6Coefficient of Variation
8
Mar 13, 2026
4.2Coefficient of Variation
8
Mar 13, 2026
5,228Average Returns
8
Mar 13, 2026