Check out our latest project ✨ OpenChapter.io: free ebooks the way its meant to be 📖

Godot Native Reinforcement Learning

An asset by minigraphx
The page banner background of a mountain and forest

Install Asset

Install via Godot

To maintain one source of truth, Godot Asset Library is just a mirror of the old asset library so you can download directly on Godot via the integrated asset library browser

Install Manually

You could also download the zip archive of the asset and manually install it into your addons/ folder

Report comment

Reason for reporting:
Character limit is 200 characters

Delete Comment

Are you sure you want to delete this Comment? This action can't be undone.

Godot Native Reinforcement Learning thumbnail image
Godot Native Reinforcement Learning thumbnail image
Godot Native Reinforcement Learning thumbnail image
Godot Native Reinforcement Learning hero image

Quick Information

0 ratings
Godot Native Reinforcement Learning icon image
minigraphx
Godot Native Reinforcement Learning

Reinforcement learning for Godot 4.5+ with NATIVE ncnn inference— statically linked C++, no C#/.NET, no external runtime.Train with the standard godot-rl (godot_rl_agents) Python stack; deploy native on web/WASM, console, mobile, desktop, and edge — targets an ONNX/.NET pipeline can't reach.▶ TRY IT IN YOUR BROWSER (no install): https://minigraphx.github.io/godot-native-rl/The demo launcher as a single-threaded WASM build. Trained agents run on native ncnn with no Python at runtime.Pick "Evolution Lab" and the browser tab TRAINS a neural net in front of you.▶ Watch a demo: https://youtu.be/Cud1gvbjg0IWHAT YOU GET• A GDExtension that runs trained policies natively via ncnn — one NcnnRunner node, no runtime dependency.• A godot_rl_agents-compatible training bridge: train with the stock godot-rl Python package, convert to ncnn, deploy.• Declarative reward authoring (Signal→Reward + RewardBuilder) and a sensors library (raycast, relative-position, camera/pixels, grid, navmesh, entity/attention).• In-engine ES training (ESTrainer) — evolve a policy with NO Python, NO socket, NO backprop; runs on every target including web.• Deploy extras a Python-server framework can't match: web/WASM, INT8 quantization, async + batched crowd inference, LOD policy switching, recurrent/LSTM state, continuous + multi-discrete actions.EXAMPLES (all ship a trained net and run standalone — browse them in the launcher above or the examples/ folder)Chase the target • Seek & Avoid • Rover 3D • Ball Chase (SAC) • Fly By (PPO continuous) • Hide & Seek (multi-policy self-play) • Coop Collect (MA-POCA) • Quadruped Walk / Hurdles / Jump • Hexapod • 3DBall & GridWorld (Unity-parity) • Visual Chase (CNN from pixels) • Memory Chase (LSTM) • plus the live Evolution Lab.INSTALLEnable the plugin, then open the demo launcher (F5).Full guide:https://github.com/minigraphx/godot-native-rl/blob/main/docs/guide/getting-started.mdLINKS• Source & README: https://github.com/minigraphx/godot-native-rl• Guides (getting started, running examples, training, deploying, sensors): https://github.com/minigraphx/godot-native-rl/tree/main/docs/guide• ncnn vs ONNX Runtime — honest decision guide: https://github.com/minigraphx/godot-native-rl/blob/main/docs/ncnn_vs_onnx.md• Issues / questions: https://github.com/minigraphx/godot-native-rl/issuesPre-1.0: the API and wire protocol may still change between minor versions.MIT licensed.You don't need a license to run Models in your games.Plugin zip link: https://github.com/minigraphx/godot-native-rl/releases/download/v0.4.0/godot-native-rl-addon-v0.4.0.zipDemos zip link: https://github.com/minigraphx/godot-native-rl/releases/download/v0.4.0/godot-native-rl-examples-v0.4.0.zip

Supported Engine Version
4.5
Version String
0.4.0
License Version
MIT
Support Level
community
Modified Date
22 hours ago
Git URL
Link to Git URL
Issue URL
Link to Issues URL

Godot Native RL

Native ncnn inference for Godot 4.5+. Train with godot_rl_agents, deploy a statically-linked C++ GDExtension (no .NET, no Python runtime) on desktop, mobile, console, and web. This package bundles the GDScript library and prebuilt NcnnRunner binaries for every platform.

Quick start

  1. Enable the plugin: Project → Project Settings → Plugins → Godot Native RL. (Also auto-packs your model files into exports — see below.)
  2. Add a trained ncnn model (*.ncnn.param + *.ncnn.bin) to your project.
  3. Drive an agent with NcnnAIController2D / NcnnAIController3D (control_mode = NCNN_INFERENCE, plus the model_param_path / model_bin_path exports), or use NcnnRunner directly:
var runner := NcnnRunner.new()
runner.input_blob_name = "in0"
runner.output_blob_name = "out0"
# Load from bytes so it works inside an exported .pck (incl. web), not just the editor.
runner.load_model_from_buffers(
    FileAccess.get_file_as_bytes(param_path),
    FileAccess.get_file_as_bytes(bin_path))
var action := runner.run_discrete_action(PackedFloat32Array(obs))

Exporting your game

With the plugin enabled, your *.ncnn.param / *.ncnn.bin are packed into exports automatically. (Godot otherwise skips these raw data files and the game fails at runtime with cannot read model files — on every platform.)

Web: in the Web export preset set Extension Support: ON and Thread Support: OFF. The bundled WASM binary is single-threaded, so the game needs no COOP/COEP headers — it runs on itch.io / GitHub Pages with no server configuration.

Full documentation

Converting trained models, training backends, INT8 quantization, sensors, and the API reference: https://github.com/minigraphx/godot-native-rl

Reinforcement learning for Godot 4.5+ with NATIVE ncnn inference
— statically linked C++, no C#/.NET, no external runtime.

Train with the standard godot-rl (godot_rl_agents) Python stack; deploy native on web/WASM, console, mobile, desktop, and edge — targets an ONNX/.NET pipeline can't reach.

▶ TRY IT IN YOUR BROWSER (no install): https://minigraphx.github.io/godot-native-rl/
The demo launcher as a single-threaded WASM build. Trained agents run on native ncnn with no Python at runtime.
Pick "Evolution Lab" and the browser tab TRAINS a neural net in front of you.

▶ Watch a demo: https://youtu.be/Cud1gvbjg0I

WHAT YOU GET
• A GDExtension that runs trained policies natively via ncnn — one NcnnRunner node, no runtime dependency.
• A godot_rl_agents-compatible training bridge: train with the stock godot-rl Python package, convert to ncnn, deploy.
• Declarative reward authoring (Signal→Reward + RewardBuilder) and a sensors library (raycast, relative-position, camera/pixels, grid, navmesh, entity/attention).
• In-engine ES training (ESTrainer) — evolve a policy with NO Python, NO socket, NO backprop; runs on every target including web.
• Deploy extras a Python-server framework can't match: web/WASM, INT8 quantization, async + batched crowd inference, LOD policy switching, recurrent/LSTM state, continuous + multi-discrete actions.

EXAMPLES (all ship a trained net and run standalone — browse them in the launcher above or the examples/ folder)
Chase the target • Seek & Avoid • Rover 3D • Ball Chase (SAC) • Fly By (PPO continuous) • Hide & Seek (multi-policy self-play) • Coop Collect (MA-POCA) • Quadruped Walk / Hurdles / Jump • Hexapod • 3DBall & GridWorld (Unity-parity) • Visual Chase (CNN from pixels) • Memory Chase (LSTM) • plus the live Evolution Lab.

INSTALL
Enable the plugin, then open the demo launcher (F5).

Full guide:
https://github.com/minigraphx/godot-native-rl/blob/main/docs/guide/getting-started.md

LINKS
• Source & README: https://github.com/minigraphx/godot-native-rl
• Guides (getting started, running examples, training, deploying, sensors): https://github.com/minigraphx/godot-native-rl/tree/main/docs/guide
• ncnn vs ONNX Runtime — honest decision guide: https://github.com/minigraphx/godot-native-rl/blob/main/docs/ncnn_vs_onnx.md
• Issues / questions: https://github.com/minigraphx/godot-native-rl/issues

Pre-1.0: the API and wire protocol may still change between minor versions.
MIT licensed.
You don't need a license to run Models in your games.

Plugin zip link: https://github.com/minigraphx/godot-native-rl/releases/download/v0.4.0/godot-native-rl-addon-v0.4.0.zip
Demos zip link: https://github.com/minigraphx/godot-native-rl/releases/download/v0.4.0/godot-native-rl-examples-v0.4.0.zip

Reviews

0 ratings

Your Rating

Headline must be at least 3 characters but not more than 50
Review must be at least 5 characters but not more than 500
Please sign in to add a review

Quick Information

0 ratings
Godot Native Reinforcement Learning icon image
minigraphx
Godot Native Reinforcement Learning

Reinforcement learning for Godot 4.5+ with NATIVE ncnn inference— statically linked C++, no C#/.NET, no external runtime.Train with the standard godot-rl (godot_rl_agents) Python stack; deploy native on web/WASM, console, mobile, desktop, and edge — targets an ONNX/.NET pipeline can't reach.▶ TRY IT IN YOUR BROWSER (no install): https://minigraphx.github.io/godot-native-rl/The demo launcher as a single-threaded WASM build. Trained agents run on native ncnn with no Python at runtime.Pick "Evolution Lab" and the browser tab TRAINS a neural net in front of you.▶ Watch a demo: https://youtu.be/Cud1gvbjg0IWHAT YOU GET• A GDExtension that runs trained policies natively via ncnn — one NcnnRunner node, no runtime dependency.• A godot_rl_agents-compatible training bridge: train with the stock godot-rl Python package, convert to ncnn, deploy.• Declarative reward authoring (Signal→Reward + RewardBuilder) and a sensors library (raycast, relative-position, camera/pixels, grid, navmesh, entity/attention).• In-engine ES training (ESTrainer) — evolve a policy with NO Python, NO socket, NO backprop; runs on every target including web.• Deploy extras a Python-server framework can't match: web/WASM, INT8 quantization, async + batched crowd inference, LOD policy switching, recurrent/LSTM state, continuous + multi-discrete actions.EXAMPLES (all ship a trained net and run standalone — browse them in the launcher above or the examples/ folder)Chase the target • Seek & Avoid • Rover 3D • Ball Chase (SAC) • Fly By (PPO continuous) • Hide & Seek (multi-policy self-play) • Coop Collect (MA-POCA) • Quadruped Walk / Hurdles / Jump • Hexapod • 3DBall & GridWorld (Unity-parity) • Visual Chase (CNN from pixels) • Memory Chase (LSTM) • plus the live Evolution Lab.INSTALLEnable the plugin, then open the demo launcher (F5).Full guide:https://github.com/minigraphx/godot-native-rl/blob/main/docs/guide/getting-started.mdLINKS• Source & README: https://github.com/minigraphx/godot-native-rl• Guides (getting started, running examples, training, deploying, sensors): https://github.com/minigraphx/godot-native-rl/tree/main/docs/guide• ncnn vs ONNX Runtime — honest decision guide: https://github.com/minigraphx/godot-native-rl/blob/main/docs/ncnn_vs_onnx.md• Issues / questions: https://github.com/minigraphx/godot-native-rl/issuesPre-1.0: the API and wire protocol may still change between minor versions.MIT licensed.You don't need a license to run Models in your games.Plugin zip link: https://github.com/minigraphx/godot-native-rl/releases/download/v0.4.0/godot-native-rl-addon-v0.4.0.zipDemos zip link: https://github.com/minigraphx/godot-native-rl/releases/download/v0.4.0/godot-native-rl-examples-v0.4.0.zip

Supported Engine Version
4.5
Version String
0.4.0
License Version
MIT
Support Level
community
Modified Date
22 hours ago
Git URL
Link to Git URL
Issue URL
Link to Issues URL

Open Source

Released under the AGPLv3 license

Plug and Play

Browse assets directly from Godot

Community Driven

Created by developers for developers