Multi-Agent Reinforcement Learning on Simple Spread N=6 benchmark leaderboard · SOTA2 Research

Multi-Agent Reinforcement Learning on Simple Spread N=6

0.0107CollisionsMAPPO

Evaluation Results

Method	Links
MAPPO2026.06		0.0107	0.0007	2.82
MAPPO2026.06		0.0174	0.0012	2.88
MAPPO2026.06		0.0183	0.0012	7.23
MAPPO-Lag2026.06		0.0188	0.0013	4.48
QMIX2026.06		0.0217	0.0014	1.85
QMIX2026.06		0.0243	0.0016	1.8
MAPPO-Lag2026.06		0.031	0.0021	7.37
QMIX2026.06		0.0313	0.0021	2.88
IQL2026.06		0.0802	0.0053	12.87
IQL2026.06		0.096	0.0064	14.21
IQL2026.06		0.1058	0.0071	15.83
DCG2026.06		0.1425	0.0095	24.17
DCG2026.06		0.1458	0.0097	24
DCG2026.06		0.1544	0.0103	24.17
CG-CMARL2026.06		1.2	0.08	7.33
CG-CMARL2026.06		1.53	0.1	11.75
CG-CMARL2026.06		1.72	0.11	22.38
CG-CMARL2026.06		2.89	0.19	28.67
CG-CMARL2026.06		2.99	0.2	33.21
CG-CMARL2026.06		3.9	0.26	34.83
CG-CMARL2026.06		4.13	0.28	33.29
CG-CMARL2026.06		4.28	0.29	33.92
CG-CMARL2026.06		4.57	0.3	33.75