This directory contains examples that demonstrate how to use the Inference Gateway with Docker Compose.
- Docker
- Docker Compose
- Basic - Simple setup with a single model provider
- MCP - Model Context Protocol integration with multiple tool servers
- Hybrid - Configuration with multiple model providers (cloud and local)
- Tools - Function calling and tool usage examples
- Authentication - Adding authentication to your gateway
- Monitoring - Metrics collection and visualization with Prometheus and Grafana
Each example directory contains:
- A README with specific instructions
- A
docker-compose.ymlfile - An
.env.examplefile
To run any example:
- Navigate to the example directory:
cd examples/docker-compose/[example-name]- Copy the environment file and customize as needed:
cp .env.example .env- Start the services:
docker compose up -d- Follow the specific instructions in the example's README for testing and usage
Common environment variables used across examples:
| Variable | Description | Default |
|---|---|---|
SERVER_PORT |
Port the gateway listens on | 8080 |
LOG_LEVEL |
Logging level | info |
PROVIDER_*_API_KEY |
API key for specific providers | - |
If you encounter any issues with these examples, please open an issue on GitHub.