Docling

open-source

Open-source document parsing for gen AI, from IBM Research

Visit site How we handle links

About

Docling parses PDF, DOCX, PPTX, XLSX, HTML, EPUB, images and more, with page layout, reading order, table structure, formulas and OCR for scanned PDFs. It exports Markdown, HTML, DocTags and lossless JSON, runs locally for sensitive or air-gapped data, and integrates with LangChain, LlamaIndex, CrewAI and Haystack. It also ships an MCP server and an API server. The code is MIT-licensed; models carry their own licenses. Started at IBM Research Zurich and hosted by the LF AI & Data Foundation.

Is this your tool?

Make sure builders find it. See Stackdrop Featured.

Feature this tool: $49/month Questions? Ask us

Verified listing, a labeled spot in your category, a pinned week in New Drops, a newsletter mention and a monthly clicks report. Cancel anytime.

Claim this listing

Compare Docling

PDF to Markdown for LLMs

The Drop

Get new AI tools for builders every week. Free.

One short email. Unsubscribe anytime.

Related tools

Atelier

Dev tools

The studio your AI coding agents work in

paid#mcp#claude-code

CodeRabbit

Dev tools

AI code reviews

freemium#code-review#github

EwwCode

Dev tools

Social media for code

free#community#code-review

Graphicle

Dev tools

Code graph and memory for coding agents

open-source#rust#mcp

Hivelane

Dev tools

Air traffic control for your coding agents

paid#coding-agents#orchestration

Hugging Face

Dev tools

The AI community hub

freemium#models#datasets