R3F
HERO DEMOCircuit Racing
Train and save neural drivers, race their frozen checkpoints, or take the wheel. Two licensed 3D vehicles, three laps and shared arcade physics.
Imitation MLP · DQN · Double DQNView →
Read the tutorial →Train RL agents in your browser
IgnitionAI is the ML-Agents of the JavaScript creative ecosystem. Describe your game world, call env.train(), and deploy anywhere via ONNX. No Python, no server, no GPU cluster.
$npm install @ignitionai/core @ignitionai/backend-tfjs @ignitionai/environments
Quick Start
Train your first agent in 7 lines
No neural network code. No hyperparameter tuning. No config files. Describe your world, call train(), and the framework does the rest.
cartpole.ts
import { IgnitionEnvTFJS } from '@ignitionai/backend-tfjs';
import { CartPoleEnv } from '@ignitionai/environments';
const cartpole = new CartPoleEnv();
const env = new IgnitionEnvTFJS(cartpole);
env.train('dqn'); // Zero config. It just works.
// env.infer(); // Switch to inference after training.
// env.setSpeed(50); // Turbo training (50x faster).That's it. The agent learns. The pole stays up.
Built for creative JS devs
Reinforcement learning that fits your stack
Describe your world in a class, call env.train(), and watch your agent learn in real time. No Python, no server, no GPU cluster.
Everything you need to ship RL.
Three algorithms, auto-configuration, ONNX export, HuggingFace storage, and 8 working demos. All TypeScript, all open source.
Train → ONNX → deploy anywhere
Train in the browser with TensorFlow.js, then export to ONNX and deploy in Unity (Sentis), Unreal (NNE), Python, C++, or edge devices. One pipeline from prototype to production.
Three algorithms, one API
DQN, PPO, and tabular Q-Learning. Switch with one word: env.train('ppo'). Override hyperparameters only when you need fine control.
R3F-first
Pair it with your Three.js scene — training loop runs independently of the render loop.
Auto-configuration
Input size deduced from observe(), action size from actions.length. You never touch network code.
WebGPU acceleration
TensorFlow.js auto-selects WebGPU → WebGL → WASM → CPU. No CUDA, no install.
Turbo training
env.setSpeed(50) — 50× faster convergence. Drop back to 1× to watch the agent play.
HuggingFace storage
Save and load trained models from HF Hub with one line. Share weights like a dataset.
Strict TypeScript
No any, Zod-validated configs, 184+ tests, CI/CD. Production-ready from day one.
8 working demos
Circuit Racing, Maze, CartPole, MountainCar, CartPole 3D, Drone Navigation, Maze 3D, GridWorld. Clone, run, learn.
Interactive Playground
Train an agent in your browser
Pick an algorithm, hit Train, and watch the reward climb in real time. This is a simulation — open a live demo below to run real RL.
1x
Ready
Click Train to start
Episode 0
Episode 50
Reward 200
Live Demos
Watch agents learn in real time
8 interactive environments showing IgnitionAI in action. Each runs in your browser. Circuit Racing supports keyboard play against learned drivers; the other environments explore reinforcement learning.
R3F · Rapier
Maze
A low-poly humanoid trains in a Rapier 3D arena with prefab walls, sensor rays, doors, and moving hazards.
Q-Table · DQN · PPOView →
Read the tutorial →2D Canvas
CartPole
Classic pole-balancing benchmark with Euler physics.
DQN · PPOView →
Read the tutorial →2D Canvas
MountainCar
Sparse reward challenge. Explore the momentum needed to reach the hilltop.
DQN · PPOView →
Read the tutorial →R3F
CartPole 3D
Classic CartPole rendered in 3D with React Three Fiber. Metallic materials and shadows.
DQN · PPOView →
Read the tutorial →R3F · Physics
Drone Navigation
A quadcopter learns to fly to moving target points. Rigid-body physics, 8 thrust combos, real gravity and torque.
DQNView →
Read the tutorial →R3F
Maze 3D
A robot navigates a 3D maze to find keys, avoid traps, and reach the exit. Raycast observations, pathfinding challenge.
DQN · Q-TableView →
Read the tutorial →2D Grid
GridWorld
A discrete navigation environment for exploring rewards and learned action values.
Q-Table · DQN · PPOView →
Read the tutorial →Recent updates
What's new
IgnitionAI ships fast. Here's the latest release — everything else lives on the full changelog.
v0.2.0-devLanding, Docs, and Physics DemoApril 15, 2026
- Landing & sitePublic site live at [ignitionai.dev](https://ignitionai.dev)
- Landing & siteSingle Next.js deployment serving the landing, the docs, and the demos
- Landing & siteBrand: animated flame GIF logo, indigo palette, 3 custom inline SVG feature diagrams (ONNX pipeline, algorithm cards, R3F integration)
- Landing & siteFlame favicon across every page (`app/icon.svg`) and every demo
- Documentation20-page Nextra documentation site under `/docs`
Open source · MIT licensed
Ready to train your first agent?
One install command, one TrainingEnv class, one call to env.train(). Your agent learns in your browser, then deploys anywhere via ONNX.