How MarkdownPDF compares
Where it is the right tool, where it is not, and what to use instead. Strengths of the alternatives are stated as plainly as ours.
Against typical online converters
| MarkdownPDF | Typical online converter | |
|---|---|---|
| Where the file is processed | In your browser, on your device | Uploaded to the provider's servers |
| Price | Free, no page limit | Free tier with page or size caps |
| Account | None | Often required past a few files |
| Scanned PDFs (OCR) | Included, 7 languages | Often paid plans only |
| Multi-column reading order | Gutters detected, columns read in order | Varies |
| Tables | Rebuilt from text positions; CSV export | Varies |
| Check against the original | Side by side, click a line to find it | Rarely |
| Watermark | Never | Free tiers sometimes add one |
Other tools worth knowing
Open source · Python
Marker
Better at: Deep-learning layout analysis and OCR; writes equations as LaTeX and copes well with complex scientific layouts. Runs on your own machine, so files stay private too.
Trade-off: You install Python and its models; a GPU makes it much faster. Better for a pipeline than for one quick file.
Open source · Python
Docling
Better at: Layout and table-structure models that rebuild difficult tables well, with Markdown and JSON output designed for document pipelines.
Trade-off: A developer tool: installation, models to download, and code to write around it.
Commercial · cloud
Mathpix
Better at: Excellent recognition of mathematics and scientific documents, with LaTeX and Markdown output.
Trade-off: A paid service that processes documents on its servers.
Open source · command line
pandoc
Better at: The reference for converting between Markdown, Word, HTML, LaTeX and dozens of other formats.
Trade-off: It does not read PDF as an input, so it is a complement for the next step rather than an alternative.
Use MarkdownPDF when
- the document is confidential and must not be uploaded;
- you want a result in seconds, without installing anything;
- you are on a phone or a locked-down work computer;
- the PDF is mostly text: reports, contracts, papers, forms, scans of typed pages.
Use something else when
- equations must come out as LaTeX — Marker or Mathpix;
- you process thousands of files in a pipeline — Marker or Docling;
- tables are irregular, nested or full of merged cells — Docling;
- the scan is handwritten — none of these will be reliable.
The fairest test is your own document. Try the converter, or run one of the real examples first — each lists its rough edges.