This is the GitHub organization Docling open-source project. We like to get continuous feedback from the community: take the poll!

Docling is our main open-source package. It is a powerful library which simplifies document processing, parsing diverse formats — including advanced PDF understanding — and providing seamless integrations with the gen AI ecosystem.

We support an amazing community which helps us driving forward the adoption of Docling. Give it a try and join the community!

The key repositories of Docling are:

• docling - The home of the main docling package.
• docling-core - The definition of types, transforms, serializers, etc. If it has to do with the DoclingDocument you will find it here.
• docling-parse - The backend PDF parser used by Docling.
• docling-serve - The FastAPI wrappers for running Docling as REST API and distribute large jobs.
• docling-ibm-models - The AI models powering Docling.
• docling-sdg - Synthetic data generation (SDG) on documents for dataset generation for RAG, finetuning, etc.
• docling-mcp - The definition of tools with the Model Context Protocol for document conversion, manipulation and generation agents.
• docling-java - A Java API for interacting with Docling, currently based on docling-serve.

Docling is hosted as a project in the LF AI & Data Foundation.

The project was started by the AI for knowledge team at IBM Research Zurich.