01
Infer the structure
The model separates document fields, repeating rows, totals, and notes based on the file in front of it.
How does local PDF to Markdown work?
The local PDF to Markdown tool reads selectable PDF text in your browser and rebuilds headings, paragraphs, lists, and simple tables as editable Markdown. It does not upload the document or call an AI model.
What to expect
Text-based PDFs can be converted locally up to the displayed file and page limits. Scanned pages need an OCR tool because they contain pixels rather than selectable text.
The full workflow
The output stays beside the evidence.
Check the proposed fields against the source, correct a value, then export the reviewed editable Markdown.
Dynamite Docs- 01Read embedded PDF text
- 02Check reading order
- 03Download Markdown
02
Correct against the source
The source remains visible while you edit cells, rename columns, inspect warnings, and mark the result reviewed.
03
Send useful data onward with Dynamite Docs
Download editable Markdown, or move a recurring workflow into Google Sheets, Drive, webhooks, or the API.
Dynamite Docs- Headings
- # Title and ## Section
- Paragraphs
- Reading-order text with blank lines
- Lists
- - Bullets and 1. numbered items
Choose the AI route
Bring a free provider key. Keep the review workflow.
This PDF to Markdown tool runs locally and needs no AI key. For scans, photos, and structured extraction, open AI & Processing, choose Connect Provider, get a key from the provider, paste it once, and verify it.
Your provider key is encrypted before it is stored.
Switch between hosted, own-key, privacy-restricted, and local models.
The same source-first review screen works whichever route you choose.
Choose the page for the job using Dynamite Docs
Use PDF to Markdown when your PDF contains selectable digital text and you need clean, editable Markdown for documentation, technical notes, knowledge bases, or LLM context. Before processing, verify that the text can be highlighted in a browser or PDF viewer, confirming the document has an embedded text layer rather than scanned images.
After conversion, check heading hierarchy from # through ###, review list nesting and paragraph flow, and confirm that page headers and footers have been cleaned cleanly. For multi-column layouts, read through the generated Markdown continuously to ensure the reading order follows the logical narrative flow rather than jumping between columns.
If your document is a scan, photo, or screenshot without embedded text, use Image to Markdown, which applies OCR. When structured data fields and tables take priority over narrative prose, use PDF to JSON or the PDF Table Extractor. This local browser conversion requires no cloud uploads and consumes 1 PE per converted page on all plans.
Local PDF to Markdown uses no AI model. Every user spends 1 PE per successfully converted page. The browser parser accepts text-based PDFs up to 50 MB and 500 pages.
Setup, review and export questions
Does the PDF leave my browser?
No. For this local workflow the browser reads the file and runs the PDF parser in a dedicated worker. It does not upload the PDF to an extraction endpoint or AI provider.
Can it convert a scanned PDF?
Not with the local parser. A scanned page is an image and has no useful text layer. If the parser finds no readable text, choose an OCR-capable model explicitly.
Will tables and columns be exact?
Simple tables and single-column pages usually produce the cleanest Markdown. Spanning cells, dense columns, formulas, charts, and unusual reading order can need manual edits.
Does local PDF to Markdown use my Free allowance?
Yes. Each successfully converted page costs 1 PE for Free and anonymous users. The browser sends only the page count, file hash, and idempotency key to the usage ledger. It never sends the PDF or Markdown text.
What should I review after local conversion?
Read the full output in order. Check heading levels, list nesting, paragraph breaks, columns, quotations, repeated headers, footnotes and simple tables. Compare any missing symbol or unusual character with the PDF, especially when the file uses a custom embedded font.
Process the whole folder in one run.
Starter and above add stored files, batch jobs and connected exports. Email intake requires Pro or Scale. File and processing limits depend on your plan.
Compare batch plans