Godot RL Agents

Create smarter NPCs and game characters using state-of-the-art reinforcement learning

Watch Demo Get Started

Watch how AI agents learn to navigate complex environments, avoid obstacles, and achieve objectives through reinforcement learning.

Key Features

🎮

Reinforcement Learning

Train agents using state-of-the-art RL algorithms.

🤖

Multi-Agent Training

Enable agents to learn from each other and collaborate.

🎲

Game Integration

Seamlessly integrate with popular game engines.

👁️

Visual Training

Watch your agents learn in real-time with visual feedback.

🌳

Behavior Trees

Create complex agent behaviors using visual editors.

🚀

Export & Deploy

Export trained models for production deployment.

Training Approaches

Parallel Training

Train multiple agents simultaneously in isolated environments to accelerate learning

Layered Training

Build complex behaviors by training agents in progressively challenging scenarios

Multi-Agent Training

Enable agents to learn from each other and develop cooperative or competitive strategies

Quickstart Guide

GitHub Repository

View on GitHub

Godot RL Agents

A library for training agents in the Godot game engine using state-of-the-art deep reinforcement learning algorithms.

Language
Python
License
MIT
Quick Install
        pip install godot-rl

Please ensure you have successfully completed the quickstart guide before following this section.

Godot RL Agents supports 4 different RL training frameworks, the links below detail a more in depth guide of how to use a particular backend:

Installation and First Training

Installation

Install the Godot RL Agents library. We recommend using a virtual environment (venv or Conda).

        pip install godot-rl

Download Examples

Download one or more examples, such as BallChase, JumperHard, FlyBy.

        gdrl.env_from_hub -r edbeeching/godot_rl_JumperHard

Set Permissions

Add run permissions on the game executable.

        chmod +x examples/godot_rl_JumperHard/bin/JumperHard.x86_64

Train and Visualize

Start training with visualization enabled.

        python examples/stable_baselines3_example.py --env_path=examples/godot_rl_JumperHard/bin/JumperHard.x86_64 --experiment_name=Experiment_01 --viz

Video Tutorial

This video tutorial demonstrates how to install the library and create a custom environment.