Welcome to LLM Lounge — a self-contained guide to understanding and running large language models on your own hardware. This started as a set of notes for our team and grew into a full curriculum covering everything from your first local model conversation through agents, custom tooling, and hybrid cloud workflows.
The ten modules here follow the actual learning path we found most useful: get a model running in under 10 minutes, learn how to pick the right one for your hardware and task, understand what’s happening under the hood (tokens, transformers, quantization, memory), then move into practical use — chat interfaces, documents, automation, agents, coding harnesses, and combining local and cloud models. The final modules cover advanced setup, media generation, and where the field is heading. Each module is self-contained but builds on the previous ones.
Everything here is grounded in real hardware. We cover what actually runs on Apple Silicon Macs, what’s feasible on a mid-range PC, and where the tradeoffs lie. The goal is not to overwhelm you with theory but to give you the mental models and practical knowledge to make good decisions when choosing models, tools, and workflows for your own projects.