This guide helps you deploy a local Large Language Model (LLM) server on your Apple MacBook (Intel CPU or Apple Silicon (M-series)) with a user-friendly chat interface. By running LLMs locally, you ensure data privacy, improve reliability with offline capabilities, and leverage cutting-edge tools for efficient AI workflows
- macOS 11.0 or later (Intel, Apple Silicon (M-series))
- At least 8 GB of RAM (16 GB recommended for optimal performance)
- Admin privileges to install software
- Python
- Ollama
- Llama 3.2 1B
- Open WebUI
Homebrew is a free and open-source software package management system that simplifies the installation of software on Apple's operating system, macOS.
Follow steps from: https://brew.sh
Python is a high-level, general-purpose programming language.
brew install [email protected]Note: Install Python 3.11, which is required to run Open WebUI. Use Homebrew for an easy installation process.
Ollama is a lightweight, extensible framework for building and running language models on the local machine.
brew install ollamaNote: At the time of writing, I am running version 0.5.4
Llama (Large Language Model Meta AI, formerly stylized as LLaMA) is a family of autoregressive large language models (LLMs) released by Meta AI.
ollama run llama3.2:1bOpen WebUI is an extensible, self-hosted AI interface that adapts to your workflow, all while operating entirely offline.
python3.11 -m venv venv
source venv/bin/activate
python3 -m pip install open-webuiNote: At the time of writing, I am running version 0.5.2
open-webui serveNavigate to http://localhost:8080 and select "llama3.2:1b" from the drop down.
Congrats! You now have a local LLM deployment. Welcome to your own personal AI assistant.