docling
Get your documents ready for gen AI
Docling is an MCP server that parses and converts documents from multiple formats into AI-ready structured content. It handles PDFs with advanced understanding of layout, tables, and formulas, as well as DOCX, PPTX, XLSX, HTML, images, audio, video, and many other formats. The server provides seamless integration with AI agents through the Model Context Protocol, enabling LLMs to extract and work with document content. Developers use it to feed clean, structured document data into generative AI workflows via LangChain, LlamaIndex, Crew AI, Haystack, and MCP-compatible clients.
Key Features
Use Cases
- 01Converting research papers and technical PDFs into Markdown for RAG pipelines and LLM context
- 02Extracting structured data from business documents, financial reports (XBRL), and patent filings (USPTO) for AI analysis
- 03Processing scanned documents and images with OCR to make legacy content accessible to AI agents
- 04Transcribing audio and video files into text for content indexing and search applications
- 05Feeding email archives and presentations into knowledge bases for AI-powered document Q&A
- 06Building document processing workflows in agentic AI frameworks like LangChain and LlamaIndex
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
docling — FAQ
What is the Docling MCP server?+
Docling is an MCP server that converts documents from 20+ formats (PDF, DOCX, images, audio, video, etc.) into structured content that AI agents can consume. It provides advanced PDF understanding, OCR, and export to Markdown, JSON, and other formats through the Model Context Protocol.
How do I install the Docling MCP server?+
Install Docling via pip using 'pip install docling', then configure your MCP client to launch it. Python 3.10 or higher is required; the package works on macOS, Linux, and Windows for both x86_64 and arm64 architectures.
Which AI clients work with Docling?+
Docling works with any MCP-compatible client and has native integrations with LangChain, LlamaIndex, Crew AI, and Haystack. It can also run as a standalone API server (docling-serve) for broader integration scenarios.
Do I need API keys to use Docling?+
No API keys are required for core document processing. Docling runs locally and supports air-gapped environments. If you use optional Visual Language Models or ASR features, model-specific requirements may apply.
Is Docling free and open source?+
Yes, the Docling codebase is released under the MIT license and is free to use. Individual models integrated with Docling may have their own licenses; refer to the original model packages for details.
What formats can Docling convert?+
Docling supports PDF, DOCX, PPTX, XLSX, HTML, EPUB, ODF (ODT, ODS, ODP), images (PNG, JPEG, TIFF), audio (WAV, MP3), video (MP4, AVI, MOV, MKV, WebM), email (EML, MSG), LaTeX, plain text, Markdown, and specialized formats like XBRL and USPTO patents.
How do I install docling?+
Open the source repository on GitHub and follow its README. docling is a mcp server — MCP Agents Market links you directly to the official repo.
Is docling free?+
docling is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.