This FastAPI application provides an API for the RAG (Retrieve, Augment, Generate) system, allowing users to ask questions and receive generated responses.
pip install fastapi uvicorn langchain langchain-community fastembed chromadb
To get the server running locally:
- Clone the repository
- Install the requirements:
- Install Ollama, Run Ollama.
- Run the development server:
uvicorn main:app --port 8086 --reload
POST /ask: Submit a question and receive a response from LLM.POST /feed/add: Stores data to chroma vector store.POST /feed/update: Update data stored in chroma vector store.POST /feed/delete: Delete data stored in chroma vector store.POST /reset: Reset the chroma vector storage.POST /all-docs: Lists all documents stored in vector storage.