Customised RAG
Local CPU-based LLaMA model with retrieval for accurate docu
Choose to summarize text or answer questions from context