Learn local AI, one step at a time
10 bite-sized lessons with animations and a quick question at the end of each. No jargon you haven’t been introduced to. About 37 minutes from zero to running your own model.
0/10
Start lesson 1
10 short lessons · about 37 minutes in total
Part 1
The basics
What local AI is and what a model needs from your computer.
- 1What is local AI?Start hereRunning an AI model on your own computer instead of in someone else’s cloud. 3 min
- 2Memory: where the model livesA model has to sit in memory while it runs, and what kind of memory decides how fast it is. 4 min
- 3How big is a model?What “7B” or “35B-A3B” means, and how size turns into gigabytes. 4 min
- 4Quantization: shrinking modelsHow a model gets 3–4× smaller with only a small loss in quality. 3 min
Part 2
What makes it run well
Why one computer flies and another crawls.
- 5Why “it fits” isn’t enoughA model can fit in memory and still make your computer miserable to use. 3 min
- 6Context: the model’s short-term memoryHow much text a model can keep in mind at once, and what that costs. 4 min
- 7Speed: tokens per secondThe numbers that decide whether a model feels instant or sluggish. 4 min
- 8Chat vs. coding agentsWhy a model that’s great for chat can be painfully slow as a coding agent. 4 min
Part 3
Getting set up
The software involved, and your first model.