LMOS-Config

Configuration management and validation for LMOS (Language Model Orchestration System)

Overview

LMOS-Config provides a unified configuration system for the entire LMOS ecosystem. It includes schema validation, IDE support via JSON schema, and flexible loading options for both local and HTTP-based configurations. S3 integration planned.

Features

• 🔍 Schema Validation: Pydantic-based validation for all configuration options
• 💡 IDE Integration: VSCode integration with JSON schema for autocompletion
• 🌐 Flexible Loading: Support for both local and HTTP-based configuration files
• 🔄 Auto-generated Schema: GitHub Actions workflow to maintain up-to-date JSON schema
• 📝 YAML & JSON Support: Load configurations from either YAML or JSON formats

Usage

Basic Configuration

Add the schema reference to your YAML config file to enable auto complete:

# yaml-language-server: $schema=https://lmos-io.github.io/LMOS-config/schema.json

internal_configuration:
  redis:
    redis_url: redis://localhost:6379
  # ... more configuration

2. Loading Configuration

Local Configuration

Set LMOS_CONFIG_YAML_PATH env var to the path for your config.yaml IE: "path/to/your/config.yaml"

HTTP Configuration

Set LMOS_CONFIG_HTTP_URL env var to the url for your config.yaml IE: "https://example.com/config.yaml"

Configuration Structure

Internal Configuration

• Redis: Connection settings for Redis
• Database: Database connection configuration
• Prometheus: Logging and monitoring settings

Services Configuration

• Router: LMOS Router settings
• LLM Runner: Language model service configurations
• STT Runner: Speech-to-text service configurations
• TTS Runner: Text-to-speech service configurations
• Rerank Runner: Document reranking service configurations

Development

Prerequisites

# Install dependencies
pip install -r requirements.txt

Generate Schema

# Generate new schema.json
python export-schema.py

Running Tests

# Test local configuration loading
python quick_test_local.py

# Test HTTP configuration loading
python quick_test_http.py

Continuous Integration

The repository includes a GitHub Actions workflow (schema_gen.yml) that:

1. Generates the JSON schema from the Pydantic models
2. Deploys the schema to GitHub Pages
3. Makes it available at https://lmos-io.github.io/LMOS-config/schema.json

Project Structure

├── lmos_config/
│   ├── ConfigTypes/
│   │   └── base.py          # Core configuration models
│   ├── http_config.py       # HTTP configuration loader
│   └── local_config.py      # Local configuration loader
├── example_configurations/   # Example configuration files
├── export-schema.py         # Schema generation script
└── .github/workflows/       # CI/CD workflows

Contributing

0. Join the discord, chat about what you'd like to do
1. Fork the repository
2. Create a feature branch
3. Submit a pull request

License

Apache 2.0 - See LICENSE for details

Part of the LMOS Ecosystem

This package is part of the larger LMOS (Language Model Orchestration System) ecosystem. Visit LMOS.io to learn more about our other packages and services.