The RAG Chatbot architecture consists of the following components:

• Document Processor: Handles the ingestion and processing of documents to extract and store embeddings.
• Embedding Generator: Generates vector embeddings for processed documents to facilitate efficient retrieval.
• QA Pipeline: Manages the retrieval of relevant documents and generates responses using a language model.
• API Interface: Provides endpoints for interacting with the chatbot, including querying and receiving responses.

You can watch the video of testing this application clicking on the preview below: