Quick Start

Get OpenMLR running in under 2 minutes.

Choose Your Path

I want to...	Method	Time
Try it out quickly	Docker (Production)	2 min
Deploy to cloud	Render / Heroku	5 min
Develop locally	Local Development	5 min

---

Docker (Production)

The fastest way to run OpenMLR. Uses pre-built images from Docker Hub.

git clone https://github.com/xprilion/OpenMLR.git
cd OpenMLR
cp .env.example .env
docker compose -f docker-compose.prod.yml up -d
# or: make up

Open http://localhost:3000 and create your account.

No API keys needed to start

OpenMLR has an onboarding flow that guides you through adding API keys after you log in. You can configure everything from Settings > Providers in the UI.

What's running

Service	Port	Description
Web app	3000	FastAPI backend + React frontend
Worker	-	Celery background job processor
PostgreSQL	5433	Database
Redis	6380	Job queue + pub/sub

Commands

make up       # Start production stack
make logs     # Watch logs
make down     # Stop all services

---

Cloud Deployment

One-click deploy to Render or Heroku:

After deployment, add your LLM API keys via the platform's dashboard or through the OpenMLR Settings UI.

---

Local Development

For contributing or customizing OpenMLR.

Prerequisites

Tool	Version	Install
Python	3.12+	python.org
uv	latest	curl -LsSf https://astral.sh/uv/install.sh | sh
Node.js	20+	nodejs.org
pnpm	9+	npm i -g pnpm

Setup

git clone https://github.com/xprilion/OpenMLR.git
cd OpenMLR
make install              # Install Python + Node dependencies
cp .env.example .env      # Create config file
make infra                # Start PostgreSQL + Redis in Docker
make db-fresh             # Create database tables
make dev                  # Start dev servers

Open http://localhost:5173 (Vite dev server with hot reload).

Service	Port	URL
Web UI	5173	http://localhost:5173
API (Swagger)	3000	http://localhost:3000/docs
Docs site	4000	http://localhost:4000 (if running make docs-dev)

See Setup & Installation for detailed options.

---

First Steps

1. Create account — First user is automatically registered
2. Add API keys — Go to Settings > Providers and add at least one LLM key (OpenAI, Anthropic, or OpenRouter)
3. Start chatting — You're in Plan mode by default. Use Cmd+B / Cmd+E to switch between Plan and Execute modes.

---

Next

• Configuration — Environment variables and settings
• Agent Harness — How the agent works
• Modes — Plan vs Execute mode explained