Chat with PDF using Gemini 💁

A RAG (Retrieval-Augmented Generation) app that lets you upload PDF files and ask natural language questions about their content using Google Gemini.

---

How It Works

PDF → Extract Text → Split into Chunks → Generate Embeddings → Store in FAISS → User asks Question → Retrieve Relevant Chunks → Gemini generates Answer

---

Tech Stack

• Python
• Streamlit
• LangChain
• FAISS
• Google Gemini API
• PyPDF2

---

Project Structure

rag-document-qa/
├── app.py
├── requirements.txt
├── .env
├── .gitignore
└── README.md

---

Setup

1. Clone the repository

git clone https://github.com/Vishalkrishna3434/rag-document-qa.git
cd rag-document-qa

2. Install dependencies

pip install -r requirements.txt

3. Add your API key

Create a .env file in the project folder:

GOOGLE_API_KEY=your_api_key_here

Get your free API key from: https://aistudio.google.com/app/apikey

4. Run the app

streamlit run app.py

---

Usage

1. Upload one or more PDF files from the sidebar
2. Click Submit & Process and wait for "Done"
3. Type your question in the text box
4. Get your answer

---

Models Used

Purpose	Model
Embeddings	models/gemini-embedding-001
Answer Generation	gemini-2.5-flash

---

Requirements

• Python 3.9 – 3.12
• Google Gemini API key (free at AI Studio)

---

Example Use Cases

• Research paper Q&A
• Notes summarization
• Document search
• Study assistant
• Legal / technical document analysis

---

Security

• Never commit your .env file
• faiss_index/ is generated locally — do not push to GitHub
• If your API key is accidentally pushed, regenerate it immediately at AI Studio

---

Known Limitations

• Requires internet connection
• Large PDFs may take a few seconds to process
• FAISS index resets if the folder is deleted

---

Future Improvements

• Chat history across questions
• Support for DOCX and TXT files
• Cloud deployment
• Persistent vector database