Docling: Streamline Document Processing for Generative AI Applications

Summary

Docling is a powerful Python library designed to simplify document processing, preparing diverse formats for generative AI applications. It offers advanced parsing capabilities, including sophisticated PDF understanding, and provides a unified document representation. With seamless integrations into the AI ecosystem, Docling empowers developers to build robust AI solutions.

Use at your own risk

OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of code from these repositories is the user's own responsibility. Always review the repository, source code, dependencies, licenses, and security implications before running or installing anything. OSRepos is not responsible for issues, damages, or losses resulting from third-party repositories.

Introduction

Docling is an open-source project aimed at simplifying document processing, making documents ready for generative AI applications. It excels at parsing diverse document formats, including advanced PDF understanding, and offers seamless integrations with the generative AI ecosystem. With Docling, you can transform complex documents into structured, usable data for AI models.

Installation

Installing Docling is straightforward using pip:

pip install docling

Note: Python 3.9 support was dropped in Docling version 2.70.0. Please use Python 3.10 or higher.

Docling works on macOS, Linux, and Windows environments for both x86_64 and arm64 architectures. For more detailed installation instructions, refer to the official documentation.

Examples

Convert a Document (CLI)

You can convert a document directly from the command line:

docling https://arxiv.org/pdf/2206.01062

This generates a .md file in the current directory containing structured document content.

You can also use Visual Language Models (VLMs) like GraniteDocling via the CLI:

docling --pipeline vlm --vlm-model granite_docling https://arxiv.org/pdf/2206.01062

Python Usage (Recommended)

For programmatic integration, Python usage is recommended:

from docling.document_converter import DocumentConverter

source = "https://arxiv.org/pdf/2408.09869"  # a document via a local path or URL
converter = DocumentConverter()
result = converter.convert(source)
print(result.document.export_to_markdown())  # output: "## Docling Technical Report[...]"

More advanced usage and configuration options are available.

Why Use Docling?

Docling offers a robust set of features that make it an essential tool for document processing and AI integration:

Multi-Format Support: Parses a wide range of formats, including PDF, DOCX, PPTX, XLSX, HTML, EPUB, email formats, images, and more.
Advanced PDF Understanding: Goes beyond basic extraction, understanding page layout, reading order, table structure, code, formulas, and image classification.
Unified Representation: Provides a unified, expressive DoclingDocument representation format for easy manipulation.
Plug-and-Play Integrations: Seamlessly connects with popular AI frameworks like LangChain, LlamaIndex, Crew AI, and Haystack for agentic AI.
Local Execution: Ensures data privacy and security with local execution capabilities, ideal for sensitive data and air-gapped environments.
Comprehensive OCR Support: Includes extensive OCR support for scanned PDFs and images.
Flexible Services: Can be run as a service with the API server (docling-serve) or connected to any agent using the MCP server.