docling-project/docling

Project overview

It addresses a concrete bottleneck in generative AI pipelines—preparing diverse document formats for use—with advanced PDF understanding, multi-format export, local execution for sensitive data and air-gapped environments, and an easy setup path from pip install to a generated Markdown file.

Project type

Document Processing

Deployment

Refer to project documentation

License

MIT

Best for

• Developers who need to prepare diverse document formats for generative AI pipelines, including PDFs requiring layout, reading order and table structure understanding.
• Teams working with sensitive data or in air-gapped environments that need local execution capabilities.

Key capabilities

• Parses multiple document formats including PDF, DOCX, PPTX, XLSX, HTML, EPUB, images, audio and email formats, with advanced PDF understanding including page layout, reading order and table structure.
• Exports documents to various formats including Markdown, HTML, WebVTT, DocLang, DocTags and lossless JSON.

Getting started

• Run `pip install docling`, then `docling https://arxiv.org/pdf/2206.01062`; a .md file is generated in the current directory.

Evidence and sources

• README: from docling.document_converter import DocumentConverter
• README: Docling simplifies document processing by parsing diverse formats — including advanced PDF understanding — and providing seamless integrations with the generative AI ecosystem.
• README: 🗂️ Parsing of [multiple document formats][supported_formats] including PDF, DOCX, PPTX, XLSX, HTML, EPUB, Apple Pages, Numbers & Keynote, WAV, MP3, WebVTT, Box Notes, email format…
• README: docling https://arxiv.org/pdf/2206.01062
• README: converter = DocumentConverter() result = converter.convert(source)

Source updated at: 2026-10-23T18:23:54Z

Open original GitHub project