About MKD
What this is
MKD is a free file-to-Markdown converter with one job: making documents cheap enough to actually use with AI. PDFs, Word files, slide decks, spreadsheets, images and whole ZIP archives go in; lean, structured Markdown comes out — the same content at a fraction of the tokens an AI chat would charge for the original file.
Who makes it
MKD is an independent project, built and run by its maker rather than a company with a sales team. That shapes the product more than any feature list: there is no account system because nothing needs you identified, no premium tier because nothing is held back, and no roadmap of upsells because the converter on this page is the whole product.
Why it's free — honestly
Conversion runs entirely in your browser, so each additional user costs close to nothing to serve — there is no per-file server bill that would force metering. The site is funded by a single, clearly-labelled ad. That's the entire business model, stated plainly: your files are never the product, because we never even see them.
The privacy stance
Every conversion happens on your device. The parsing engines — for PDFs, Word documents, slides, spreadsheets and OCR — are loaded into your browser and run there; your file is never uploaded, stored or logged. This isn't a policy promise that requires trusting us; it's how the software is built, and you can watch the network tab while converting to confirm it. The details live in the privacy policy.
How we write about it
The claims on this site cite their sources — token figures come from vendor documentation, not invented percentages — and every example conversion shown on the format pages is genuine output from this converter, produced by running a real file through it. Where the tool has limits (scanned PDFs, old binary formats, handwriting), the pages say so rather than overselling. The blog follows the same rule.
Contact
Questions, bug reports, or a file that converted badly? Write to hello@markdownlite.com. Reports of documents that convert poorly are especially welcome — the parsers improve by meeting real files.