README.md

---
title: Unified Document Extraction API
emoji: 📄
colorFrom: blue
colorTo: indigo
sdk: docker
app_file: app.py
pinned: false
---

# 🚀 Unified Document Extraction API

**One API, Two Engines: Docling + DocStrange**

Extract structured data from any document using AI-powered engines.

## Features

- ✅ **Docling** - Advanced document parsing with structure preservation
- ✅ **DocStrange** - GPU-accelerated intelligent document processing
- ✅ **Multiple formats** - PDF, DOCX, XLSX, PPTX, Images, and more
- ✅ **Structured output** - Markdown, JSON, Tables

## API Endpoints

- `GET /` - Health check
- `GET /engines` - List available engines
- `POST /convert` - Full document conversion
- `POST /convert/markdown` - Markdown only
- `POST /convert/tables` - Tables only

## Usage

```bash
# Convert with Docling
curl -X POST "https://YOUR_SPACE.hf.space/convert?engine=docling" \
  -F "file=@document.pdf"

# Convert with DocStrange
curl -X POST "https://YOUR_SPACE.hf.space/convert?engine=docstrange" \
  -F "file=@document.pdf"
```

## Integration

Works with **DataSync** application for ERPNext integration.

## License

MIT