Anydoc

A simple AI-powered document analyzer that extracts summaries, key insights, and answers questions from your PDFs and text files.

🚀 Try Anydoc

Live Demo: https://anydoc.onrender.com

Upload a document and instantly get AI-powered summaries, key insights, and answers to your questions.

⭐ If you find Anydoc useful, consider giving the project a star on GitHub!

What it does

• Smart Summaries - Get the gist of any document in a few sentences
• Key Insights - Pull out the most important points without reading everything
• Q&A - Ask questions about your document and get straight answers
• Dark Mode - Easy on the eyes (because we all stare at screens too much)
• Fast - Uses Groq's speedy API so you're not waiting around
• Works on mobile - Because sometimes you need to check documents on your phone

The Tech Stack

• Frontend: Streamlit
• AI: Groq with Llama 3.1
• PDF Processing: PDFPlumber
• Language: Python

Getting Started

What you need

• Python 3.8 or newer
• A Groq API key (grab one for free at console.groq.com)

Local Setup

1. Clone the repo

   git clone https://github.com/So-coder-ai/Anydoc.git
   cd Anydoc

2. Install the requirements

   pip install -r requirements.txt

3. Set up your API key

   cp .env.example .env

   Edit .env and add your Groq API key.

4. Run it

   streamlit run app.py

5. Open your browser

   Go to:

   http://localhost:8501
   

Project Structure

anydoc/
├── app.py              # The main Streamlit app
├── utils/
│   ├── parser.py       # Handles PDF and text file reading
│   └── llm.py          # Talks to Groq API
├── requirements.txt    # Python packages we need
├── .env.example        # Example environment file
├── .gitignore          # Stuff git should ignore
└── README.md           # This file

Configuration

Environment Variables

Create a .env file with:

GROQ_API_KEY=your_groq_api_key_here
GROQ_MODEL=llama-3.1-8b-instant

Note: If you are using a different Groq model in your deployment, update GROQ_MODEL accordingly.

File Support

• PDF files (as long as they have selectable text)
• Plain text files (.txt)

Note: Scanned PDFs won't work because the text needs to be selectable, not just an image.

Contributing

Found a bug or have an idea? Great!

1. Fork this repo

2. Create a feature branch

   git checkout -b feature/your-cool-idea

3. Commit your changes

   git commit -m "Add your cool idea"

4. Push to your branch

   git push origin feature/your-cool-idea

5. Open a Pull Request

License

MIT License - do whatever you want with it.

Credits

• Streamlit for making web apps simple to build
• Groq for the fast AI API
• PDFPlumber for handling PDFs

Got Problems?

• Open an issue for bugs
• Open an issue for feature requests
• Email: shivnarainofficial24@gmail.com

---

🌐 Live Demo: https://anydoc.onrender.com 💻 GitHub: https://github.com/So-coder-ai/Anydoc