← All tools

Convert

PDF → Markdown

Batch-converts PDFs into clean Markdown, with an AI pass to fix what rules miss.

In progressUpdated 12 Jul 2026
The converter drop zone, with the rule-based and Gemini-assisted processing engines selectable.
The converter drop zone, with the rule-based and Gemini-assisted processing engines selectable.
PDF → Markdown screenshots

Why it exists

Design and research work arrives as PDFs — specs, audits, vendor docs — and none of it is greppable, diffable, or pasteable into notes. Copy-paste out of a PDF gives you broken line wraps, orphaned headings, and tables flattened into word soup.

This converts a whole batch at once. A rule-based parser does the structural pass, then an optional AI refine step cleans up the headings and tables the rules get wrong. Images are extracted and referenced rather than dropped.

Built withFastAPIGeminiReactVitepdf.js

Changelog

    • Batch upload — queue several PDFs, track per-file status.
    • Two-stage parsing: rule-based structural pass, then optional Gemini refine per file.
    • Sequential processing lock so a large batch cannot overload CPU or blow the API quota.
    • Image extraction with static serving, so Markdown keeps working references.