This repository provides concise PyTorch implementations of major multi-agent reinforcement learning (MARL) algorithms including MAPPO, MADDPG, MATD3, QMIX, and VDN. The implementations are designed to work with benchmark environments like the Multi-Agent Particle-World Environment (MPE) and StarCraft Multi-Agent Challenge (SMAC). The code supports both discrete and continuous action spaces, with modifications to the MPE source code to facilitate easy switching between action space types.