In this talk, I'll give a quick introduction to LLM and how to use it in a PHP application. I'll show some examples using the LLPhant project including a retrieval-augmented generation (RAG) system using a local LLM (Llama 3) and Elasticsearch as a vector database.