Convert DOCX to MD
Word to Markdown, in your browser.
Drop your .docx file
or click to browse · max 50MB
or paste with ⌘V
Converting DOCX to Markdown extracts the structure and prose of a Word document into the plain-text format that powers GitHub READMEs, Obsidian notes, static-site blogs, and most modern note-taking tools. Headings become # lines, lists become - lines, links become [text](url). The visual styling of the original Word document is dropped; what survives is the content in a form that's portable, diffable, and version-controllable.
When you'd use it
Move docs into git
Engineering teams want documentation in repositories where it can be reviewed, diffed, and versioned. Converting Word drafts to Markdown lets you drop them into a GitHub or GitLab repo as-is.
Migrate to Obsidian or Notion
Knowledge-management tools like Obsidian, Logseq, and Notion all import Markdown cleanly but choke on docx. Converting first preserves your headings and structure on the way in.
Publish to a static blog
Writing in Word but publishing on a Markdown-based blog (Astro, Hugo, Jekyll, MDX-powered Next sites)? Converting to Markdown skips the painful manual reformatting step.
How it works
-
Your file never leaves the browser.
-
Conversion runs locally via WebAssembly and Canvas.
-
Download. No account, no history.
Technical notes
This converter uses mammoth.js's Markdown export, which maps Word styles to GitHub-flavored Markdown. Headings (H1–H6), paragraphs, ordered and unordered lists, bold, italic, and hyperlinks all round-trip cleanly. Tables convert to Markdown table syntax (| col | col |). Images extract as base64 data URIs in the output, so the Markdown stays self-contained — useful for archival but verbose; for production publishing you may want to extract images to separate files.
Frequently asked
What happens to my DOCX files?
Nothing leaves your browser. The conversion runs locally on your device using WebAssembly and the Canvas API. No file is uploaded to a server — including ours.
Is the converted MD identical to the original?
MD is a lossless format. Pixel data round-trips exactly when the input is also lossless.
Is there a file size limit?
There is no hard cap, but very large files (hundreds of MB) may be slow on older devices because everything happens in your browser's memory.
Do you collect data on my file or my use?
Your file is invisible to us — it never leaves your browser. For aggregate page-view counts we use Cloudflare's server-side edge logs (no JavaScript beacon runs in your browser). We display Google AdSense ads on conversion result pages; the AdSense script loads on every page and may set advertising cookies after you accept the consent banner (Google Consent Mode v2 asks first). The ads see the page URL and standard browsing signals — they never see your file, because the file never leaves your device.
Will my Word formatting come through?
Structurally, yes — headings, lists, tables, bold, italic, and links convert cleanly. Visual styling (font choice, color, custom margins, page breaks) is dropped because Markdown has no equivalent. That's the point of the format: structure preserved, presentation flexible.
What about images?
Images are embedded as base64 data URIs directly in the Markdown. The output file is portable and self-contained, but larger than Markdown referencing external image files. For a publishing workflow, you'll likely want to extract images and rewrite the references — that's a manual step we don't automate today.
Does this support Word's footnotes / endnotes?
Mammoth converts footnotes to Markdown footnote syntax where it can. Complex Word features (track changes, comments, embedded objects) are dropped — they have no clean Markdown equivalent.