Turn a PDF into useful, editable content.

Faithful document reconstruction from PDF to structured Markdown.

Drop a PDF hereor use Choose PDFPDF files up to your configured limit · nothing is uploaded

Import workspace Open Markdown

Offline-first Checkpointed Export-ready GitHub

PDF

Choose another PDF

01 Document

All pagesCustom

Page rangeUse commas and ranges

02 Extraction Text tablesRecover structured table content Editable equationsDetect mathematical expressions Preserve visual contentKeep figures and formula fallbacks OCR when neededRead pages without usable text

03 Processing Remove repeated headers Remove repeated footers Join split paragraphs Detect headings Normalize task lists Preserve flow notices Keep unresolved visuals Keep source page markers

04 AdvancedOCR and technical limits

Force OCRApply OCR to every selected page

OCR languageRender DPI

PDF password

Max file (MB)Max pages

Checkpoint everyPages per processing batch Strict modeStop on the first page error

MuPDF WebAssembly and the bundled English OCR model run locally. Bounded batches release memory between checkpoints.

Workspace dataCheckpoint controls

To clear the app’s full local cache, use Reset local data on the welcome screen.

Ready to extract0%

ReadyConfigure the document, then start extraction.0%

⌘ FChanges save automatically

Page

Detail

Results

Quality overview

Waiting

Pages processed: 0

Word settingsOptional

Document titleSource page breaks

Export locallyNo upload