Image to Markdown with OCR

Extract the text from screenshots, photos of documents and scans. PNG, JPG, TIFF and WebP images are read with OCR in 11 languages and returned as Markdown text.

  • PNG, JPG, TIFF, WebP
  • 11 OCR languages
  • Scanned PDFs too

What the image converter keeps

  • Text: recognized with the Tesseract OCR engine at 300 DPI.
  • Languages: English, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Turkish, Russian and Ukrainian, also combined (for example eng+deu in the API).
  • Paragraphs: lines are grouped into paragraphs.
  • Scanned PDFs: work the same way: use the PDF converter.

Example output

A screenshot of a short notice becomes plain Markdown text:

Office closed on Friday

The office will be closed on Friday, 14 November, for maintenance.
Remote work is available as usual. Urgent requests: facilities@example.com

How to convert Image to Markdown

  1. Drop a PNG, JPG, TIFF or WebP image above.
  2. OCR runs on the server; it takes a few seconds.
  3. Copy the recognized text. Check names and numbers: OCR is very good but not perfect.

Image to Markdown API

Set ocrLanguages=eng+deu for mixed-language documents. OCR pages per request depend on the plan (5 on the free plan, up to 500).

curl -X POST "https://anytomarkdown-pdf-to-markdown-and-document-parser-api.p.rapidapi.com/v1/convert/raw?ocrLanguages=eng" \
  -H "X-RapidAPI-Key: YOUR_RAPIDAPI_KEY" \
  -H "X-RapidAPI-Host: anytomarkdown-pdf-to-markdown-and-document-parser-api.p.rapidapi.com" \
  -F "file=@scan.png"

Get a free API key API overview

What OCR can and cannot do

OCR recognizes printed text well, especially in screenshots and clean scans. Results get worse with handwriting, very small or low-contrast text, and photos taken at an angle. Tables in images come out as text lines rather than Markdown tables; if you have the original document, convert that instead.

Frequently asked questions

Which languages are supported?

English, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Turkish, Russian and Ukrainian. The free tool uses English; the API accepts any combination.

Can it read handwriting?

Not reliably. The OCR engine is made for printed text.

Can I convert a scanned PDF?

Yes, use the PDF to Markdown converter. Pages without text are recognized automatically.

Related converters

PDF to Markdown

Headings, tables and lists from any PDF, with OCR for scans.

Open converter →

Word (DOCX) to Markdown

DOCX, DOC, RTF and ODT with styles, lists, tables and footnotes.

Open converter →

Email (EML, MSG) to Markdown

EML and Outlook MSG, with attachments converted too.

Open converter →

URL to Markdown

Any web page as clean Markdown, without menus or ads.

Open converter →