Multi-Agent Reinforcement Learning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
45.06Average Episodic Return
8
Feb 26, 2026
31.02Average Episodic Return
8
Feb 26, 2026
43.1Average Episodic Return
8
Feb 26, 2026
91.9Win Rate
8
Apr 2, 2026
98.9Win Rate
8
Apr 2, 2026
90.9Win Rate
8
Apr 2, 2026
83.926h_vs_8z Win Rate
8
Feb 26, 2026
20Average Score
8
Feb 26, 2026
19.71Averaged Score
8
Feb 26, 2026
20.9Averaged Score
8
Feb 26, 2026
20.94Averaged Score
8
Feb 26, 2026
19.4Averaged Score
8
Feb 26, 2026
20.94Test Win Rate
8
Feb 26, 2026
97Test Win Rate
8
Feb 26, 2026
100Win Rate
7
Jun 30, 2026
100Win Rate
7
Jun 30, 2026
100Win Rate
7
Jun 30, 2026
98.4Win Rate
7
Jun 30, 2026
100Win Rate
7
Jun 30, 2026
100Win Rate
7
Jun 30, 2026
100Win Rate
7
Jun 30, 2026
100Win Rate
7
Jun 30, 2026
100Win Rate
7
Jun 30, 2026
100Win Rate
7
Jun 30, 2026
3.8Interactions to 80% Reward (x10^5)
7
Jun 25, 2026