# edbeeching/godot_rl_agents

An Open Source package that allows video game creators, AI researchers and hobbyists the opportunity to learn complex behaviors for their Non Player Characters or agents

Repository: https://github.com/edbeeching/godot_rl_agents
Canonical: https://ross.abutalabs.com/products/godot_rl_agents
Language: Python
License: MIT
License Family: permissive
Topics: reinforcement-learning, simulation, godot
Last push: 2026-07-10T20:19:17+00:00

## Health v2 (maintenance only)
Score: 64/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 91, release rhythm 8, longevity 100
- inputs: {"age_days": 1889, "days_push": 54, "days_rel": 554, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1575, forks 116 (observed 2026-08-28T04:05:05.871547+00:00)

## What it is
Godot RL Agents is an open-source Python package that bridges games built in the Godot Engine with reinforcement learning algorithms, enabling training of complex NPC and agent behaviors. It wraps four popular RL frameworks (StableBaselines3, Sample Factory, Ray RLLib, CleanRL) and supports 2D/3D games, memory-based agents, and AI sensors.

## Use cases
- train NPCs to learn complex behaviors in Godot games
- run reinforcement learning experiments in 2D and 3D game environments
- use StableBaselines3 or RLLib to train agents in a Godot simulation
- build memory-based agents with LSTM or attention for game AI
- research RL in custom game environments
- teach game characters to jump, chase, or fly via RL

## When to choose
- you build games in Godot and want learned agent behaviors
- you want an open-source, MIT-licensed RL-to-game-engine bridge
- you want to experiment with multiple RL frameworks against the same game environment
- you are an AI researcher needing a free simulation environment for RL

## When to avoid
- your game engine is Unity or Unreal rather than Godot
- you need production game AI without a Python training pipeline
- you want supervised learning or LLM-based agents rather than reinforcement learning

## Facets
- artifact type: library
- maturity: active
- function: reinforcement-learning, machine-learning, simulation, game-engine
- domain: reinforcement-learning, machine-learning, simulation
- platform: python, cross-platform, windows
- tags: godot, game-ai, npc-behavior, stable-baselines3, rllib, sample-factory, cleanrl, lstm-agents, game-development, linux, macos

## Member repositories
- edbeeching/godot_rl_agents (main) score 64

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:05.871547+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:57:35.637343+00:00, confidence not recorded.
  - readme: https://github.com/edbeeching/godot_rl_agents (fetched 2026-08-28T04:05:05.871547+00:00, sha de3586aec062)
- Data as of 2026-08-30T08:39:29.467469+00:00.