FullMoon is an iOS application designed for private, local chat with large language models (LLMs) on Apple devices. It targets iPhone, iPad, and Mac users seeking on-device AI interaction with a focus on privacy and customization, leveraging Apple's MLX framework for optimized performance on Apple Silicon.

How It Works

The app utilizes Apple's MLX Swift framework, an array framework specifically built for machine learning research on Apple Silicon. This allows FullMoon to efficiently run LLMs directly on the device, ensuring data privacy as chat history is saved locally. The architecture is optimized for Apple's hardware, promising smooth performance for supported models.

Quick Start & Requirements

• Install via the App Store.
• Requires an Apple device (iPhone, iPad, Mac) with Apple Silicon.
• Supported models include Llama 3.2 1B, Llama 3.2 3B, and DeepSeek-R1-Distill-Qwen-1.5B-4bit.

Highlighted Details

• Optimized for Apple Silicon performance.
• Local chat history storage for enhanced privacy.
• Customizable app appearance.
• Built using MLX Swift.

Maintenance & Community

Information regarding maintainers, community channels, or roadmaps is not detailed in the provided README.

Licensing & Compatibility

The README does not specify a license. Compatibility for commercial use or closed-source linking is undetermined.

Limitations & Caveats

The app currently supports a limited set of LLM models. Information on broader model compatibility, advanced customization options, or potential performance limitations on older Apple hardware is not provided.