All Tools
PDF to Markdown
Reconstruct headings, lists and paragraphs as clean Markdown.
Drop a PDF here or click to select
How it works
- Drop a text-based PDF (articles, docs, reports).
- The layout is analyzed page by page: font sizes become headings, bullet and numbered lines become lists, hyphenated line breaks are repaired.
- Review the Markdown in the preview, toggle page separators on or off.
- Copy it to the clipboard or download the .md file.
What people use it for
- Moving PDF documentation into a wiki, Obsidian, or a static-site repo.
- Preparing clean, structured text from PDFs for LLM pipelines and RAG ingestion.
- Extracting an article's structure without retyping the headings and lists.
Good to know
- Conversion is heuristic, not AI: headings are inferred from relative font size, so decorative layouts can misclassify a line — the structure is right far more often than not, but skim the output.
- Tables are detected conservatively (only clearly aligned columns become pipe tables); complex tables come out as plain lines rather than wrong tables.
- Scanned PDFs have no text layer to read — run PDF OCR first.
Frequently asked questions
Does this use an AI model?
No — it is deterministic layout analysis (font sizes, positions, spacing), which means identical input always gives identical output, instantly, and nothing is sent to any model provider.
How are multi-column PDFs handled?
Lines are read in visual order; standard single-column documents convert cleanly, while complex multi-column magazine layouts may interleave — simpler layouts give better Markdown.
Is the PDF uploaded?
No. Text extraction and the Markdown reconstruction run entirely in your browser.