---
title: Granite Document Summarization
emoji: 📚
colorFrom: blue
colorTo: gray
sdk: gradio
sdk_version: 5.25.2
app_file: app.py
pinned: false
license: apache-2.0
short_description: Summarize documents with Granite models.
---

# Granite Document Summarization

A streamlined document summarization application powered by IBM's Granite 3.3 Large Language Model.

## Features

- Upload and process documents in multiple formats (PDF, DOCX, PPTX, HTML)
- Customize summaries by specifying the number of sentences (1-10) or paragraphs (1-3)
- GPU-optimized for fast processing
- Memory-efficient chunking and processing

## How It Works

1. Document Conversion: Uploaded documents are converted to markdown using Docling
2. Text Chunking: The text is split into manageable chunks with semantic boundaries preserved
3. Embedding Creation: Text chunks are converted to embeddings using Nomic Embed
4. Summary Generation: IBM Granite 3.3 generates concise summaries based on your specifications

## Requirements

- Python 3.9+
- GPU environment for optimal performance
- Dependencies listed in requirements.txt

## Usage

1. Upload a document (PDF, DOCX, PPTX, HTML)
2. Choose between sentence-based or paragraph-based summarization
3. Adjust the length using the slider or radio buttons
4. Click "Summarize" to generate your summary

## License

Apache 2.0

---

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference