LangChain, Llama2-Chat, and zero- and few-shot prompting are used to generate synthetic datasets for IR and RAG system evaluation
Why do you think that https://github.com/leoneversberg/llm-chatbot-rag is a good alternative to LangChain-SynData-RAG-Eval