A local LLM knowledge base for humans

LLMs shouldn’t be exclusive to our Techno Feudal Lords. Running LLMs on your personal computer is actually fairly straight-forward given a little bit technical knowledge.

However, when looking around for information online, I discovered that most are lacking in actual detail, dump out half-assed articles about theoretical setups that they never tested or benchmark scenarios that are quite frankly complete bullshit. And that’s before you start to read one AI-generated article after another, that are even more void of any substance.

This is what I am trying to do here. A short, compact knowledge base and light blog about my experiences with deploying local LLMs. Written by me, a human.

Quickstart

If you have no idea about how LLMs work, what you need to know in order to run them effectively and how you can interact with them the best, check the LLM 101 page.

If you already have somewhat of an idea, check the Tools page on for concrete guidance.

If you don’t know what hardware you might need, check the Hardware page, but fair warning, this is the most opionated and least verified page.

For more references, check out the Further resources page.

What this site contains

I am trying to fill the gaps of knowledge that I have personally encountered while trying to dive into this topic. When I am writing about something here, it’s always first-hand experience. If it’s not, it is always clear, where the information is coming from.

What I am trying to put here is the practical information that are somewhat grounded to real-life usability. In other words, what can realistically be achieved with a setup that is either somewhat affordable or might already be in a lot of people’s home.

I will list other sources that I found to be very helpful here as well.

What this site doesn’t contain

I will only write here about setups I can actually verify (with the exception of hardware, I am not a millionaire unfortunately).

Therefore, you won’t giant lists of benchmarks for dozens of models, how to run a GLM 5.3 at home or what the best AI rig for $10,000 is.