LlamaChat

LlamaChat is a native macOS application that lets you chat with large language models—including LLaMA, Alpaca, and GPT4All—completely locally on your machine. It supports importing raw PyTorch model checkpoints and pre-converted .ggml model files, making model setup straightforward.

Key Features

• Local execution – All inference runs on your Mac; no cloud dependence.
• Model import – Directly load PyTorch checkpoints or .ggml files.
• Multiple model support – Works with LLaMA, Alpaca, GPT4All, and planned Vicuna support.
• Open-source stack – Built with llama.cpp and llama.swift.

Who It’s For

Developers, researchers, and AI enthusiasts who want to experiment with instruction-tuned language models locally on macOS.

Requirements

• macOS 13 (Intel and Apple Silicon).
• Users must obtain model files independently; LlamaChat does not distribute any models.