Chat with MLX

Chat with MLX is an all-in-one chat application for running large language models locally on Apple Silicon Macs using Apple’s MLX framework. It combines a chat interface, retrieval-augmented generation over your own documents and support for the growing library of MLX-converted models into a single local app, with no cloud accounts or API keys required. It is an open-source project on GitHub with around 1,600 stars.

Overview

MLX is Apple’s array framework optimized for the unified memory architecture of M-series chips, and it has become a fast way to run open models on Macs. Chat with MLX wraps MLX inference in a friendly interface: download a model, chat with it at native speed, and optionally ground answers in your own files through built-in RAG. Everything runs on-device, so conversations and documents stay private, and the app can switch between models from the MLX community hub for tasks from coding to writing.

Key Features

• Local LLM chat optimized for Apple Silicon via the MLX framework
• Built-in RAG: chat with your own documents and files
• One-click access to MLX-format models from the community hub
• Fully offline and private once models are downloaded
• Adjustable generation settings and conversation management

💳 Pricing

Chat with MLX is free and open source under the MIT license. You only need a Mac with Apple Silicon.

Who Is It For

Mac users who want a private, fast, ChatGPT-style experience with open models running entirely on their own machine.

The app is built on MLX, Apple’s machine learning framework, and on the mlx-lm tooling that converts and runs Hugging Face models in MLX format. Because Apple Silicon shares memory between CPU and GPU, even large models can run at interactive speeds on a laptop, and the project became a popular way for Mac owners to try open models locally in 2024, before dedicated desktop LLM apps became common.