RAG (Retrieval Augmented Generation) and finetuning are two popular methods for using Large Language Models (LLMs) with “custom” data. But after speaking to many customers, I noticed that it can be confusing to know which method to use, when, and why.