# Inference

<!-- https://learn-inference.com/chapters/inference -->

A generative model is trained once and then serves for the rest of its life. This chapter is about that second phase: what serving actually involves, and the three layers of engineering it takes to do well.

## 6 sections

1.  [0.1Two phases, two disciplines](https://learn-inference.com/chapters/inference/two-phases)
2.  [0.2The three layers](https://learn-inference.com/chapters/inference/three-layers)
3.  [0.3Six techniques that define the runtime](https://learn-inference.com/chapters/inference/runtime-techniques)
4.  [0.4Scale changes the problem](https://learn-inference.com/chapters/inference/scale-changes-problem)
5.  [0.5Where to put the abstraction](https://learn-inference.com/chapters/inference/abstraction)
6.  [0.6A map of what follows](https://learn-inference.com/chapters/inference/map)