notAOpenAI-LLM Custom file ---> convert into text chunks(LLM models have token size restriction) ---> LLM embedding ---> Vectors ---> vector database ---> Load Model ---> Passing Model to RetrievalQA chain