Office Document to Markdown Converter

More than 20: Word (.docx, .docm, .doc), Excel (.xlsx, .xlsm, .xls), PowerPoint (.pptx, .pptm, .ppt), PDF, Outlook messages (.msg, .eml), plus HTML, CSV, JSON, XML, EPUB, RST, RTF, TXT and ZIP. Each file is parsed on the server and comes back as clean Markdown with headings, tables and lists intact.

No. This converter reads text-based files only. It has no text recognition, so a scanned PDF or a PNG or JPG of a document has nothing in it that it can read.

Office2MD, the Windows app, has OCR built in. It reads scanned PDFs and image files on your own PC, with nothing uploaded, and handles mixed documents too: pages with real text keep it, scanned pages get recognised.

Completely free, with no account, no sign-up and no credit card. Drop a file, get Markdown back. There is no usage counter and nothing to unlock.

Tables become proper Markdown tables with pipes and dashes. Headings, bold, italic, bullet lists and numbered lists all map to their Markdown equivalents. Use the Preview tab above to see the result rendered before you copy it.

Markdown is plain text, so anything that depends on layout will not survive: charts, embedded images, text boxes, precise column widths, colours. If those matter, keep the original file as well.

No. Your file is uploaded, converted and discarded. Once the Markdown is back in your browser, nothing about the file remains on the server: it is not saved, not indexed and not passed to anyone.

If a document is confidential enough that uploading it at all is the problem, use Office2MD instead. It converts on your own PC and never touches the internet.

5 MB, which covers most documents. For anything bigger, Office2MD runs on your PC with no size limit at all, so a 300-page report is just a longer wait.

Because it is the plain-text format that AI tools, documentation systems and knowledge bases all understand. Hand a Word or PDF file to ChatGPT and the structure often gets flattened on the way in. Hand it Markdown and the headings, tables and lists arrive intact, which gives you better answers and lets you see exactly what the model received.

Outside AI, Markdown is what Obsidian, Notion, GitHub, Hugo, MkDocs and Docusaurus read natively. It is the right format for content that has to live in more than one place.

One at a time here, though a ZIP archive counts as one file and everything inside it gets converted. For real batches, Office2MD takes as many files as you drop in, of mixed types and any size, and converts them in one go.

Sort of, and not reliably. A short Word document usually comes through fine. A long Excel workbook with several sheets, or a PDF with an awkward layout, often does not: content gets summarised instead of extracted, tables come back mangled, and you cannot see what went in.

This tool does that one job properly. It extracts the content and structures it the way Markdown expects, which is also the way the models read best. Excel is where the difference shows most: every sheet is processed and labelled, rather than half-read.

All of them are converted. Each sheet becomes its own Markdown table under a heading with the sheet name, so you can always tell which numbers came from where.

No, they are an estimate. The number shown above the output assumes roughly four characters per token, which is the usual rule of thumb for English text. Every model tokenises a little differently, and code, tables and other languages skew it, so treat it as a size check rather than a billing figure.

Office2MD is a Windows app that converts 30+ formats to Markdown directly on your PC. Drop in as many files as you like, of any size, and convert them all in one go, completely offline. Every install comes with 10 free conversions, everything unlocked, so you can try it on your own documents first.