microsoft/markitdown

What is it?

Key features

• 01Converts PDF, Office documents, HTML, text formats and ZIP archives to Markdown
• 02Preserves headings, lists, tables and links in the output
• 03Provides both a command-line interface and a Python API
• 04Installs only selected format dependencies via optional extras
• 05Supports Azure Document Intelligence and Azure Content Understanding for cloud extraction
• 06Supports third-party plugins for additional formats

Use cases

• Preparing documents as Markdown context for LLM prompts or retrieval pipelines
• Batch converting Office files and PDFs for text analysis
• Extracting content from YouTube transcripts or audio files into text
• Feeding structured document content into LangChain or AutoGen workflows

Quick start

1. 1

   Install with pip including all optional dependencies

   pip install 'markitdown[all]'
2. 2

   Convert a file to Markdown from the command line

   markitdown path-to-file.pdf -o document.md

Things to know

• !Output is aimed at text analysis and LLM pipelines, not high-fidelity human-readable conversion.
• !Requires Python 3.10 through 3.14.
• !Some features such as image descriptions, Document Intelligence and Content Understanding need Azure or LLM API keys and incur costs.

Topics

FAQ

What is markitdown?

Python utility that converts PDFs, Office documents, HTML, images, audio and more into Markdown for LLM and text analysis pipelines. MarkItDown is a lightweight Python library and command-line tool that converts files such as PDF, PowerPoint, Word, Excel, HTML, CSV, JSON, XML, EPUB and ZIP archives into Markdown. It aims to preserve document structure like headings, lists, tables and links, so the output can be fed into large language models or other text analysis tools.

Is markitdown free to use? What license does it have?

markitdown is open source under the MIT license. Check the license terms before using it in commercial products.

Is markitdown still maintained?

Actively maintained. The last commit to markitdown was on Oct 4, 2026.

What can I use markitdown for?

Preparing documents as Markdown context for LLM prompts or retrieval pipelines Batch converting Office files and PDFs for text analysis Extracting content from YouTube transcripts or audio files into text Feeding structured document content into LangChain or AutoGen workflows

How do I install markitdown?

Install with pip including all optional dependencies: pip install 'markitdown[all]' Convert a file to Markdown from the command line: markitdown path-to-file.pdf -o document.md

Where can I get markitdown?

The source code and installation instructions are on GitHub: https://github.com/microsoft/markitdown