Leon is an open-source personal assistant designed to run on your server, enabling voice and text interactions with customizable "skills." It aims to consolidate various personal automation tasks into a single, extensible platform, prioritizing privacy through optional offline operation.

How It Works

Leon employs a hybrid approach, combining Large Language Models (LLMs) with simpler classification and NLP techniques for optimized speed, customization, and accuracy. The architecture includes a server, skills registry, web app, hotword detection, and inter-process communication via a TCP server and Python bridge, facilitating integration of diverse functionalities.

Quick Start & Requirements

• Install Leon CLI: npm install --global @leonle/cli
• Create Leon instance: leon create birth (stable) or leon create birth --develop (development)
• Run Leon: leon start
• Prerequisites: Node.js >= 22.13.1, npm >= 10.9.2. Supported OSes: Linux, macOS, Windows.
• Documentation: docs.getleon.ai

Highlighted Details

• Integrates transformers-based models for advanced AI capabilities.
• Features a skill registry platform for community-contributed functionalities.
• Offers optional offline mode for enhanced privacy.
• Supports voice and text interaction.

Maintenance & Community

The project is primarily maintained by Louis Grenard, with sponsorship from Appwrite and others. The project has a Discord server for community engagement and contribution.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

The development branch is noted as potentially unstable due to ongoing major changes. Documentation and README are outdated, with updates planned for the official release. The project's development pace is impacted by the lead developer's limited spare time.