edbeeching/godot_rl_agents
A reinforcement learning interface connecting the Godot game engine to Python ML backends for training AI agents in games.

Velocity · 7d
+0.8
★ / day
Trend
→steady
star history
Godot RL Agents is an open-source package that bridges the Godot game engine with machine learning frameworks, enabling developers to train agents and NPCs using reinforcement learning. It wraps four popular RL libraries (StableBaselines3, Sample Factory, Ray RLLib, CleanRL) and provides AI sensors for game observation. The toolkit supports memory-based agents with LSTM or attention mechanisms for both 2D and 3D games.