Docling
Document parsing library by IBM for converting PDFs and documents to structured data.
About
Docling, from IBM Research Zurich and now hosted by the LF AI and Data Foundation, converts documents into structured representations ready for LLM and RAG pipelines. It parses PDF, DOCX, PPTX, XLSX, HTML, EPUB, images, email formats, and even audio and video files, then exports Markdown, HTML, JSON, or its DocTags format while preserving reading order. Its models perform layout analysis, table structure recognition, and formula and code extraction, with OCR for scanned pages and speech recognition for audio inputs; the compact GraniteDocling visual language model at 258M parameters handles end-to-end conversion. Everything runs locally, which matters for sensitive documents, and an API server mode is available. Native integrations cover LangChain, LlamaIndex, Haystack, and CrewAI, plus an MCP server for agent workflows. Requiring Python 3.10 or later and MIT licensed, it is used by developers building document understanding and retrieval pipelines over messy real-world files.
Should you use Docling?
Pick it when
Pick Docling when a RAG or agent pipeline must ingest mixed PDFs, Office files, HTML, and even audio locally, under an MIT license, with ready LangChain, LlamaIndex, Haystack, CrewAI, and MCP integrations.
Look elsewhere when
Skip it if you only need tables or raw text from digital PDFs, where Camelot or PyMuPDF are far lighter. For the hardest handwritten or degraded scans, a dedicated VLM such as olmOCR or Chandra targets that problem directly.
Alternatives to Docling
- Unstructured
Lists 60+ formats with cleaning, chunking, and vector-store staging plus a paid hosted API, but its OCR leans on system tesseract rather than dedicated layout models.
- Marker
Focused PDF to Markdown with LaTeX equations and a managed service option, but its model weights need a paid license above 5 million dollars in revenue, where Docling has no cap.
- MinerU
Also parses PDFs, scans, and Office files with a scientific and technical focus, but under a custom Apache-2.0-based license with attribution terms instead of plain MIT.
- MegaParse
Optional GPT-4o or Claude vision mode for the worst layouts, but commits stopped in February 2025 and the vision path needs paid API keys.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- OCR & Document Processing
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Easy (2/5)
- License
- MIT
- Added
- Apr 3, 2026
Tags
Related Tools
Browse all OCR & Document Processing toolsMentioned in
Fine-Tuning vs RAG vs Long Context in 2026: A Decision Guide
How to choose between fine-tuning, RAG, and long context in 2026, with real costs, the failure modes of each...
Max P
PDF Parsing for RAG in 2026: MinerU, Docling, Marker Compared
A benchmarked comparison of MinerU, Docling, Marker 2, Surya, PDF-Extract-Kit and Zerox for RAG ingestion,...
Billy C