How To Use This Guide

I’ve been using ChatGPT and other Large Language Models (LLMs) to help me code for well over a year now. For the most part, it’s amazing and has helped me turn a lot of my ideas into working projects. However, there are also times when something gets updated behind the scenes, and ChatGPT loses a few IQ points. As someone who loves to tinker, I figured I’d give running local LLMs a try. Some of the advantages are allowing me to use multiple models at once, leveraging my computer’s own hardware, and have full control of my development environment.

This guide is designed to share my, and other’s, understanding of how to set up and optimize local LLMs, solely for developers looking for help with coding. There’s a lot to cover, so I’m going to focus on breaking down everything you need to know in order to pick the right model, understanding your computer’s capabilities, and benchmarking the models I use to ensure you can get the best performance possible. This guide was created from my personal notes and learnings over the past few years of coding with an LLM at my side. I’ll show you how to not only leverage LLM’s ability to generate code but also learn when and how to use it so you can scale your productivity and some tips on how to avoid relying on it like a crutch.

I hope you enjoy this guide and please feel free to contribute if you have something to add or see something that needs that needs to be changed.