This project is a Retrieval-Augmented Generation (RAG) Chatbot built with FastAPI, LangChain, ChromaDB, and Streamlit. It enables users to upload multiple documents, process them, and query for relevant responses.
Chunking - Recursive Character Splitter Embedding - sentence-transformers/all-MiniLM-L6-v2 model LLM - Groq llama3-70b-8192 model
RAG-Chatbot/
│── backend/
│ ├── chroma_utils.py
│ ├── db_utils.py
│ ├── pydantic_models.py
│ ├── main.py
│ ├── langchain_utils.py
│ ├── requirements.txt
│ ├── Dockerfile
│
│── frontend/
│ ├── api_utils.py
│ ├── streamlit_app.py
│ ├── chat_interface.py
│ ├── sidebar.py
│ ├── requirements.txt
│ ├── Dockerfile
│
│── docker-compose.yaml
│── .gitignore
│── README.md
Ensure you have the following installed:
- Docker & Docker Compose
- Python 3.11+ (if running locally)
- Groq API Key (for model inference)
- Hugging Face API Key (for embeddings)
git clone https://github.com/vasstavkumar/RAG-Chatbot.git
cd RAG-ChatbotCreate a .env file in the root directory and add the required API keys:
groq_api_key=your_groq_api_key
api_key=your_huggingface_api_keydocker-compose up --build- FastAPI will run at http://localhost:8000
- Streamlit frontend will be available at http://localhost:8501