XuanCe is an open-source ensemble of deep reinforcement learning algorithm implementations covering value-based methods (DQN, DDPG, Qmix), policy gradient approaches (A2C, PPO), and multi-agent algorithms (MADDPG, MAPPO). It provides a unified API across multiple deep learning backends and integrates with standard RL environments including gymnasium, PettingZoo, MuJoCo, and StarCraft2 for benchmarking and research.