Ollama Run Local LLMs Offline and Privately for Free

From Ollama setup to your own local AI on an 8 GB laptop with no GPU. Hugging Face, Python, RAG and embeddings
What you'll learn

---

Run powerful AI models on your own computer, with total privacy and zero API cost, even if you have never installed one
Get real results on a normal laptop with 8 GB of RAM, no dedicated GPU and no cloud bill, including Mac with Apple silicon
Build your first RAG from scratch with embeddings, so the AI answers from your own documents
Explore, choose and run open source models from Hugging Face with Python pipelines
Command Ollama from zero to advanced: install, Generate vs Chat, parameters, streaming, System Prompt and roles
Create your own custom models with a Modelfile, and get reliable answers with Structured Output and Function Calling
Understand how an LLM works inside and pick the right model for the hardware you already own
Requirements

---

A computer with at least 8 GB of RAM, 16 GB recommended, and we show you which models run well in each range
About 20 GB of free disk space to download and store the models
Basic comfort with the terminal or command line, the essentials are taught inside the course
No dedicated GPU needed, the examples run on regular laptops, including Mac with Apple silicon
No previous experience with AI or machine learning is required, we start from zero
Description

---

You heard that you can run AI on your own computer, and the first thing you thought was: that must be for people with a monster GPU and a free weekend.
It is not.
Open source models got small and good at the same time. A model that needed a data center two years ago now answers on a laptop with 8 GB of RAM, and it answers well enough for real work.
This course has a single promise: by the end of it you will have your own AI running on your own machine, private and free, and you will know exactly which tool to reach for in each situation.
The first win arrives early. In the opening lessons you already have a model downloaded and answering on your machine, with the wifi off, before you create any account or type any card number.
You will set up your Python environment, run models by task with Hugging Face pipelines, install and command Ollama from the terminal, build custom models with a Modelfile, work with embeddings, put together a basic RAG that answers from your own documents, and run multimodal models locally.
You leave with a working local setup, a map of which models fit which hardware, and the judgment to choose between a graphical interface and the terminal for each job.
I have taught more than a million students in over 200 countries, and Rodrigo Macedo builds every technical lesson with you on screen.
The window is open right now: the models are free, the tools are free, and the distance between what runs on your desk and what runs in the cloud has never been this short.
If you have a computer and an afternoon, your private AI starts running today. See you inside.
Who this course is for

---

Developers who want to add AI to their applications without paying for APIs or sending data to third parties
Professionals who work with sensitive material: proprietary code, internal documents, client information
Data professionals and machine learning engineers who want to master the open source model ecosystem
Students and AI enthusiasts who want to see how LLMs actually work by running them
Teams and companies evaluating local LLMs to cut cost and keep control of their own data
Anyone already using ChatGPT, Claude or Gemini who wants their own AI running at home
This is not for someone who wants to train or fine tune a model from scratch, here we run, customize and integrate existing open source models
This is not for someone looking for a zero setup experience, you will install tools and use the terminal
Homepage

---