Chaty Fangyuan Lin

Chaty is an AI-powered chatbot designed to run entirely locally on your device, enabling offline interactions with large language models without any reliance on cloud services or telemetry.

Key Features:

• Runs 100% locally, supporting GGUF models (Llama.cpp, Vulkan GPU) and MLX models natively on Apple Silicon.
• Includes a local coding agent that can execute tasks directly in your browser or workspace while maintaining privacy.
• Grounds answers in your own documents through a private knowledge base using RAG (Retrieval-Augmented Generation).
• Generates cited Deep Research reports by combining web search and model reasoning.
• Supports hands-free interaction via offline voice input and output, with no GPU requirements for speech processing.

Audience & Benefit:
Ideal for developers, researchers, and privacy-focused users who need a powerful AI tool that operates entirely on their machine. Chaty enhances productivity while ensuring complete control over data, as conversations and documents remain stored locally without any upload or logging to external servers.

Chaty is free and open source, with installation available via winget.