Docling
Document parsing library that converts PDFs, Word docs, and other formats into clean, structured data ready for LLM and RAG pipelines.
Overview
Docling parses complex document formats - PDFs with tables and figures, Word documents, PowerPoint files, and more - into structured representations (like Markdown or JSON) that preserve layout information an LLM can actually use, rather than the garbled plain text a naive PDF-to-text conversion produces. It's built specifically with RAG and LLM ingestion pipelines in mind.
Docling is a fit for teams feeding documents into an LLM or RAG pipeline who need accurate structure preservation, complementing Tesseract (also in this catalog) - Tesseract recognizes text from raw pixels, while Docling focuses on understanding a document's structure and layout.
- Categories
- AI & Machine Learning
- Languages
- Python
- License
- MIT
Related Projects
Haystack
Apache-2.0Open-source framework for building production-ready search and RAG (retrieval-augmented generation) applications with LLMs.
- AI & Machine Learning
Instructor
MITLibrary for getting reliable, structured data output from LLMs using function-calling APIs and Pydantic models for validation.
- AI & Machine Learning
LangChain
MITFramework for developing applications powered by large language models, agents and retrieval-augmented generation.
- AI & Machine Learning
LlamaIndex
MITData framework for connecting large language models to your own data - documents, databases, APIs - for retrieval-augmented applications.
- AI & Machine Learning
Spotted an error? Suggest an edit on GitHub.