What is pdf to markdown on Rosetta?

PDF to markdown extracts the embedded text layer from a digital PDF and returns markdown (plus per-page JSON on the API). It is PDF → Markdown, not markdown to PDF, and it is not OCR.

Does this tool support scanned PDFs or OCR?

No. Rosetta extracts text from digital PDFs with an embedded text layer only. Scanned image PDFs without a text layer are not supported.

Is there a free trial?

No free trial. You get one browser conversion without sign-in (first 5 pages, max 4.5 MB), then 200 free API credits per month when you create an API key. After that, paid plans apply.

How do I convert PDF to markdown in code?

POST to /api/v2/parse with your PDF file or URL and an X-API-Key header. See the API docs for Node, Python, and cURL examples. The MCP tool is pdf_parse.

Is there an MCP server for Cursor?

Yes. Use the remote PDF MCP server or npx -y rosetta-mcp. pdf_parse converts digital PDFs to markdown; local stdio also accepts a file path.

PDF to Markdown

PDF → Markdown · not markdown to PDF

PDF to markdown for digital PDFs with an embedded text layer. Convert one file in the browser (up to 5 pages), then use the same parse via API key or MCP — 200 free credits/month, then paid. No OCR. Not markdown to PDF.

Upload a digital PDF and get markdown back. This is pdf to markdown (PDF → Markdown), not markdown to PDF. One anonymous conversion, first 5 pages, max 4.5 MB. After that, sign up for 200 free API credits per month, then a paid plan — not a free trial. Same parse as POST /api/v2/parse and the pdf_parse MCP tool.

Digital PDFs with embedded text · no OCR · max 4.5 MB upload

Install MCP Parse API docs

Upload

Drop PDF here

Choose file

Digital PDFs with embedded text only — no OCR for scanned documents.
Max file size: 4.5 MB. Free tier: first 5 pages, 1 conversion without sign-in.

Markdown preview

Upload a PDF to see markdown output

Parse API

Same pdf to markdown endpoint your pipeline needs — POST /api/v2/parse for full documents

Parse API docs →

Install MCP

pdf_parse in Cursor, Claude Desktop, and agent workflows — remote URL or npx

PDF MCP server →

200 free credits

API key is free. 200 credits/month, then paid plans — not a free trial

Pricing →

FAQ

What is pdf to markdown on Rosetta?: PDF to markdown extracts the embedded text layer from a digital PDF and returns markdown (plus per-page JSON on the API). It is PDF → Markdown, not markdown to PDF, and it is not OCR.
Does this tool support scanned PDFs or OCR?: No. Rosetta extracts text from digital PDFs with an embedded text layer only. Scanned image PDFs without a text layer are not supported.
Is there a free trial?: No free trial. You get one browser conversion without sign-in (first 5 pages, max 4.5 MB), then 200 free API credits per month when you create an API key. After that, paid plans apply.
How do I convert PDF to markdown in code?: POST to /api/v2/parse with your PDF file or URL and an X-API-Key header. See the API docs for Node, Python, and cURL examples. The MCP tool is pdf_parse.
Is there an MCP server for Cursor?: Yes. Use the remote PDF MCP server or npx -y rosetta-mcp. pdf_parse converts digital PDFs to markdown; local stdio also accepts a file path.