Headings, lists and tables
Office documents retain their document structure where the source provides it. Merged or complex tables use HTML inside Markdown so row and column spans can survive.
HTML to Markdown
Convert local HTML or a supported document to Markdown, review the result and download the text or a ZIP with readable embedded images.
Word · Excel · PowerPoint · OpenDocument · PDF · EPUB · RTF · CSV · HTML · text · JSON · XML · Jupyter
Choose documents to begin.
Review your Markdown
Select a converted file to see its content.
Edits are included in downloads. Preview loads embedded images locally; external images stay unfetched. A ZIP keeps the Markdown and its images together.
In your browserNo uploadsNo account
How to convert HTML or documents to Markdown
- Choose local files. Select your HTML or supported documents. Conversion starts automatically; each result appears in Your files.
- Review and edit. Select a completed file, inspect its Markdown or Preview, and compare it with the source. Edits in the Markdown box are included in subsequent downloads.
- Download the result. Use Download .md for text, ZIP + images for an individual result with embedded images, or Download all as ZIP for completed files in the queue.
Document structure, tables and image files
Images travel with the text
Readable embedded document images are placed in an assets folder. Download ZIP to keep the links working. Embedded objects such as charts may need manual review.
Review before you reuse
Compare important totals, formulas, merged cells and reading order with the original. Markdown cannot reproduce every page layout. Conversion does not guarantee a pixel-perfect copy.
Spreadsheet sheets become tables. Saved formula results are read; formulas are not recalculated. Presentations become a reading sequence rather than a reproduction of slide positioning or animations.
HTML tables can remain HTML inside Markdown so cell spans are not forced into a pipe table. HTML image links are preserved but not fetched or included in ZIP files. An embedded document image and an external HTML image reference are different cases.
The Markdown output is editable. The rendered preview and an application opening the downloaded Markdown may display supported HTML, equations or relative image links differently; check the actual file and keep its assets folder with it.
Support and limits
| Source | Formats | What to check |
|---|---|---|
| Word and rich text | DOC, DOCX, DOCM, ODT, RTF | Headers, floating text boxes and embedded objects may differ. |
| Spreadsheets | XLS, XLSX, XLSM, XLSB, ODS, CSV | Sheets become tables. Formula results depend on values saved by the original application; formulas are not recalculated. |
| Presentations | PPT, PPS, POT, PPTX, PPTM, PPSX, PPSM, ODP | Slide content becomes a reading sequence; positioning and animations are not retained. |
| PDF with a readable text layer | This page does not run OCR. Scanned pages are reported as needing OCR. PDF images are not exported. Use the PDF tool for page previews and selection. | |
| Books and webpages | EPUB, HTML, HTM | EPUB images can be bundled. HTML image links are preserved but not fetched. Scripts and unsafe links are removed. |
| Text and notebooks | TXT, MD, JSON, XML, IPYNB | Use UTF-8. JSON, XML and plain text use code blocks. Notebook code is not executed; rich HTML outputs are omitted. |
Image-only files, audio, video, password-protected documents and arbitrary ZIP archives are not supported in this version. Large or damaged files may reach browser resource limits. Each conversion stops after 90 seconds; split difficult files into smaller documents.
Input limits: up to 20 files, 25 MB per file and 100 MB of source files per batch. The editable Markdown limit is 5 MB per result; converted text and image data also share a 100 MB batch budget. Oversized files or results are reported instead of offered as completed conversions.
For Word to Markdown, Excel to Markdown and PowerPoint to Markdown, choose a supported local document and review its result. For Google Docs to Markdown, download a DOCX first. This does not connect to your Google account or synchronize changes.
Compatibility with other document converters
Is this Microsoft MarkItDown running online?
No. If you are looking for MarkItDown or MarkItDown online to convert documents, this page provides a browser-local alternative for the listed formats. It uses separate conversion engines and is not affiliated with Microsoft. It does not provide every MarkItDown feature: OCR, audio and video conversion are not available.
Test files and results
Tested on: .
These four fixed English inputs were selected together in this converter and the actual Markdown and ZIP files were downloaded. The observations below describe these files, not every supported format. The input files are original site fixtures; the HTML was authored for this check. Download the input ZIP for the four original files; the check-record hashes refer to the unzipped originals.
| Source and actual download | Observed output | Boundary found or not checked |
|---|---|---|
| HTML source in input ZIP Actual HTML to Markdown result | Heading levels, bold and italic text, two list items, the link target, code indentation and the final marker were retained. The table kept its two-column span as HTML. Script text was absent from the output. | The external image stayed a reference; it was not fetched or bundled. The preview showed an image placeholder. This is not a general HTML security or layout test. |
| English DOCX source Actual DOCX to Markdown result | Three headings, a numbered list starting at 3, nested bullets, all three table data rows and a footnote were present. All table strings matched the source. | The source code block became separate text paragraphs, not a fenced code block. Check code formatting in the actual Markdown. |
| Embedded-image DOCX source Actual Markdown · Actual ZIP with images | Two PNG files were included under assets; both matched the embedded source bytes. Both relative references resolved in the preview. | The input also contains equations, which were emitted as TeX text. The preview displays that TeX as text rather than typesetting it. Mathematical equivalence and other notation were not validated. |
| Merged-cell XLSX source Actual XLSX to Markdown result | The five-row sheet kept its three merged ranges through HTML rowspan and colspan. Cell strings, amounts and prices matched after line-break normalization. | The source cell “Long description” followed by a line break and “second line” became “Long description second line”. This fixture contains no formulas; formula calculation was not tested. |
Download the actual four-file batch ZIP · Download the check record and input hashes. These files do not establish compatibility for PowerPoint, legacy Office formats, EPUB or every language. PDF-specific samples remain available in the PDF conversion test dataset; that corpus is separate from these HTML, DOCX and XLSX checks.
Local processing and privacy
Selected files are read and converted in this browser; document contents and results are not uploaded to a conversion server. Conversion components load from this site. No account or paid conversion API is required.
Readable embedded document images are stored with the result for ZIP downloads. Preview displays those images locally. External HTML image links are not fetched or bundled; following a hyperlink later, or opening downloaded Markdown in another application, can contact its destination.
Website-delivery and analytics requests are separate and described in the privacy policy; file contents and conversion results are not sent as analytics events. Use Clear to reset the queue and result on this page. Files already downloaded remain on your device.
Frequently asked questions
How do I convert a Google Doc to Markdown?
Download your document from Google Docs as Microsoft Word (.docx), then choose that local file here. Review the Markdown and select Download .md, or download ZIP to include readable embedded images. See Google’s file download instructions. This page does not connect to your Google account or open Google Docs links.
This imports an exported document into Markdown. It does not publish Markdown into Google Docs or synchronize changes with the original.
Can I convert Word DOCX to Markdown with pictures?
Yes. Headings, lists and tables are converted from the document structure. Readable embedded images are included in ZIP downloads alongside the Markdown. Floating objects and complex layouts need review.
How does Excel XLSX to Markdown handle formulas and merged cells?
Worksheets become tables. Merged cells use HTML tables inside Markdown to retain spans. Only saved formula results are read; this tool does not recalculate formulas, so a missing saved value may leave a cell empty.
Can I convert PowerPoint PPTX to Markdown?
Yes. Slide content becomes a reading sequence, and readable embedded images can be downloaded in ZIP. Slide positioning, animations and embedded charts are not reproduced as a complete slide layout.
Will HTML image links be included in my ZIP?
No. External HTML image references are preserved but not downloaded or bundled. Readable embedded images from supported document files are a separate case; see Document structure, tables and image files.
Can I convert a picture or scanned document into Markdown text?
Not in this version. Standalone images are not accepted for text recognition, and scanned PDFs need OCR. Image bundling from supported documents is separate from recognizing text in a picture.
Which languages can I convert?
Documents containing Unicode text can preserve multiple languages; this tool does not translate. PDF extraction uses PDF.js with locally hosted character maps. Simplified and traditional Chinese, Japanese and Korean PDF fixtures are tested. Correct results still depend on the source PDF's text layer and font mappings. Scans require OCR, which is not available. Arabic and Hebrew reading order and complex layouts need review. Text and HTML files must currently use UTF-8.
Page updated: