AI-summariser

Image-to-text PDF summariser with a Streamlit web UI.

Run

1. Activate your virtual environment.
2. Install dependencies:

python -m pip install transformers torch torchvision accelerate pypdfium2 pillow streamlit

3. Start the UI:

streamlit run app.py

4. In the browser:

• Upload a PDF.
• Choose how many pages to analyze.
• Click Summarize.

Notes

• First run is slower because model weights are downloaded.
• You can set HF_TOKEN to avoid Hugging Face rate-limit warnings.