[shekhar101]

Good explanation. Thanks! Can the first part, i.e. vectorizing and finding relevant chunks be done with any LLM (e.g. a self hosted one) and the second part, i.e. querying relevant chunks be done with OpenAI?