Part 2 Core domain-specific LLMs

The first three chapters in this part of the book discuss training domain-­specific SLMs. We’ll look at different approaches to SLM specialization and explore an end-to-end example of model specialization. We’ll also look at how to calculate up front how much memory you’ll need to run inference with a given SLM and how to spot areas for improvements. Chapters 5 and 6 dive deeper into SLM quantization with the ONNX framework and explore different precision formats for model quantization.