PDF GPT is a PyQt5-based desktop application that extracts text from PDF files and translates it to any target language using locally running Ollama language models. The application features a modern, intuitive interface with drag-and-drop functionality, real-time progress tracking, and seamless background processing.

• Prerequisites
• Installation
• Ollama
• Usage
• Changelog
• Contributing
• License

Before running the application, ensure you have:

• Python 3.6 or higher installed on your system
• Ollama installed and at least one model downloaded (see installation steps below)
• Required Python packages: PyPDF2, ollama, PyQt5, requests (see installation steps)

1. Visit https://ollama.ai and download Ollama for your operating system
2. Install Ollama following the instructions for your platform
3. Verify installation by opening a terminal and running:

   ollama --version

Before using the application, you need to download at least one Ollama model. Recommended models:

# Small, fast model (good for testing)
ollama pull gemma3:1b

# Medium-sized, balanced model
ollama pull llama2

# Larger, more capable model
ollama pull mistral

You can browse all available models at https://ollama.ai/library.

1. Clone the repository:

   git clone https://github.com/loyft/pdf-gpt.git
   cd pdf-gpt

2. Install required Python libraries:

   pip install -r requirements.txt

   Or install manually:

   pip install PyPDF2 ollama PyQt5 requests

Important: Ollama must be running before you start the PDF GPT application.

Option 1: Desktop Application (Recommended)

• macOS: Launch Ollama from Applications folder. It will start automatically in the background.
• Windows: Launch Ollama from the Start menu. The app runs in the system tray.
• Linux: Launch the Ollama desktop app if installed.

Option 2: Command Line

• Open a terminal and run:

  ollama serve

  Keep this terminal window open while using the PDF GPT.

To verify Ollama is running correctly, open a terminal and run:

curl http://localhost:11434/api/tags

You should see a JSON response listing your installed models. If you get a connection error, Ollama is not running.

Note: The Ollama desktop application typically starts automatically when you log in. The PDF GPT connects to Ollama at http://localhost:11434 by default.

1. Ensure Ollama is running (see Starting Ollama section above)

2. Launch the PDF GPT:

   python main.py

3. Drag and drop a PDF file into the "PDF File" area at the top of the window

4. Enter the target language (e.g., english, spanish, german, french)

5. Enter the Ollama model name (e.g., llama2, mistral, gemma3:1b)

   • Must be a model you've downloaded with ollama pull
   • The application will verify the model exists before starting translation

6. Click "Translate Text"

7. When done you can open the translated file in the same directory as the original file: originalname_targetlanguage.txt

• Improved UI/UX
• Background threading to prevent freezing issues
• Improved output file naming
• Custom model input instead of predefined (limited) dropdown menu
• Real-time progress tracking
• Improved error handling and connection verification

• Added drag-and-drop PDF functionality
• Added model selection dropdown
• File output with option to open translated file
• Basic error handling

• Initial release
• Basic PDF text extraction
• Translation using OpenAI GPT models
• Command-line interface with basic translation functionality

Contributions are welcome, please raise an issue or create a pull request. :)

This project is licensed under the MIT License - see the LICENSE file for details.