Getting started with Ooblex. Deploy in minutes with Docker.

Production-ready AI video processing platform with WebRTC streaming, edge computing, blockchain verification, and enterprise features.
View on Github Docker Hub

Quick Start with Docker Compose

Ooblex uses a microservices architecture with Docker containers for easy deployment and scaling. Get up and running in under 5 minutes with our pre-configured Docker Compose setup.

Requirements:
• Docker 24+ and Docker Compose 2.x
• 8GB RAM minimum (16GB recommended)
• NVIDIA GPU + drivers (optional, for ML acceleration)
• SSL certificates (self-signed included for dev)

Included Services:
• API Gateway (FastAPI)
• WebRTC Gateway (WHIP/WHEP)
• ML Workers (GPU-accelerated)
• Edge Computing Server
• Blockchain Service
• Redis Cache
• RabbitMQ Message Bus
• PostgreSQL Database
• Nginx Ingress
• Monitoring Stack

1. Clone and Setup

Start by cloning the repository and running the automated setup script:


# Clone the repository
git clone https://github.com/ooblex/ooblex.git
cd ooblex

# Run automated setup
./deploy.sh setup

# Or use Make
make setup

The setup script will:
• Check system requirements
• Generate SSL certificates
• Create necessary directories
• Set up environment variables
• Download ML models


# Configure environment (optional)
cp .env.example .env
nano .env

# Start all services
./deploy.sh start

# Or use Docker Compose directly
docker-compose up -d

2. Access Your Installation

Once services are running, access Ooblex at:

• Web Interface: https://localhost
• API Gateway: https://localhost:8800
• WebRTC Gateway: wss://localhost:8100
• Grafana Dashboard: http://localhost:3000
• RabbitMQ Management: http://localhost:15672

View the MJPEG test stream at: http://localhost:8081/demo.mjpg


Production Deployment Options

Kubernetes with Helm:
helm install ooblex ./charts/ooblex -f values.prod.yaml

Docker Swarm:
docker stack deploy -c docker-compose.yml ooblex

Systemd Services:
sudo cp launch_scripts/*.service /etc/systemd/system/

Cloud Providers:
• AWS EKS / ECS
• Google Cloud Run / GKE
• Azure Container Instances / AKS
• Digital Ocean Kubernetes

See our deployment guide for detailed instructions.

ML Models & Edge Modules

Ooblex includes pre-trained models and WebAssembly modules:

AI Models (ONNX format):
• Face Detection & Recognition
• Face Swap & Morphing
• Style Transfer (10+ styles)
• Background Removal
• Object Detection (YOLO)
• Super Resolution

Edge Computing (WASM):
• Browser-based face detection
• Real-time style transfer
• Background blur
• Performance monitoring

Models are automatically downloaded during setup or can be manually placed in the ml_models/ directory.

Download Models