PDF RAG Chatbot using Gemini AI
This project allows users to upload PDF files and ask questions about the content using Gemini AI and Retrieval-Augmented Generation (RAG).
Features
- Upload multiple PDFs
- Extract text from PDFs
- Chunk text intelligently
- Store embeddings using FAISS
- Ask questions from documents
- Powered by Google Gemini AI
- Deployable on Hugging Face Spaces
Tech Stack
- Streamlit
- LangChain
- Google Gemini API
- FAISS Vector Database
- Hugging Face Spaces
Setup Instructions
1. Install Dependencies
pip install -r requirements.txt
2. Add API Key
Create a .env file:
GOOGLE_API_KEY=YOUR_API_KEY
For Hugging Face Spaces:
- Go to Settings → Repository Secrets
- Add:
- Key: GOOGLE_API_KEY
- Value: Your Gemini API Key
3. Run Locally
streamlit run app.py
Deployment
Deploy easily on Hugging Face Spaces using Streamlit SDK.
Author
Built with ❤️ using Gemini AI + LangChain