Free document extraction
Extract the headings, paragraphs, tables and images of PDF files, scans, Word, PowerPoint and Excel files into Markdown and JSON.
Up to 501 pages per PDF and no monthly cap; free extractions wait in a queue. For large files or when you are in a hurry, we recommendthe OmniDoc desktop app: no page limit and no queue.
How to convert a PDF to Markdown and JSON
- 1Upload the fileSign in with an OmniDoc account, then drop the file on the area above or click “Choose file”.
- 2Start the extractionPDF files, scans and images go through layout analysis, text recognition and table recognition; Word, PowerPoint and Excel files, e-books and web pages are parsed directly.
- 3Download the resultWhen it is done, download a ZIP file with the Markdown, the JSON and the images.
Example
Page 1 of a report and the Markdown extracted from it: heading levels, paragraphs and tables are kept.
# Greener Last-Mile Delivery How five cities cut delivery emissions by a third in one year ## 1 What changed In 2026 the five pilot cities replaced diesel vans on their busiest routes with electric vans and cargo bikes, and moved sorting into micro-hubs inside the city centre. … **Table 1** Results by city, January–September 2026 | City | Parcels per day | Electric share | CO₂ per parcel (g) | |---|---|---|---| | Hangzhou | 182,000 | 74% | 96 | | Shenzhen | 236,000 | 81% | 88 | | Chengdu | 141,000 | 63% | 112 | | Xi'an | 98,000 | 58% | 121 | | Qingdao | 87,000 | 66% | 104 |
The ZIP file you download contains
document.mdThe full text as Markdowndocument.jsonStructured data in reading orderimages/The pictures in the documentmanifest.jsonA list of the files with basic information
Supported formats
| File | How it is processed |
|---|---|
| PDF, scanned PDF, PNG, JPG, WEBP and BMP images | Layout analysis, text recognition (OCR) and table recognition |
| Word、PowerPoint、Excel、CSV | The document structure is read directly |
| EPUB e-books, HTML pages | Chapters and body text are parsed |
| Markdown, TXT, SRT and ASS subtitles | Converted into the same Markdown and JSON |
Questions
Which files does the result contain?
One ZIP file with document.md (the full text as Markdown), document.json (structured data), images/ (the pictures) and manifest.json (a list of the files).
Can scans be extracted?
Yes. Scanned PDF files and images are recognised first and then organised. Check the recognised text against the original, especially numbers and tables.
What are the limits of free extraction?
Up to 501 pages per PDF and no monthly cap. Free extractions are processed in the order they are submitted, so the speed depends on the queue. We do our best on quality; if a result is not good enough, please bear with us. For large files or when you are in a hurry, we recommend the OmniDoc desktop app: no page limit and no queue.
How is this different from copying the text?
Extraction keeps the heading levels, paragraphs, tables and images and how they relate, so you get structured data rather than plain text.
Do I need an account?
Yes. Sign in with an OmniDoc account; signing up is free. Only you can see the files you upload and their results.
Where do I download the result?
Download it on this page when it is ready. The extraction runs in the background, so leaving the page does not affect it; open this page again and sign in to find earlier files under “Recent extractions”.
What if the file extension does not match the real format?
The real format is detected from the file content. For example, a .pdf file that is really a Word document is processed as .docx and comes back as .docx. Save old .doc, .xls and .ppt files as .docx, .xlsx and .pptx first.
Do I need to install anything?
No, it runs in the browser. For files with many pages or when you are in a hurry, we recommend the OmniDoc desktop app for Windows, macOS and Linux: no page limit and no queue.
The OmniDoc desktop app: smooth, 100% Office XML support, China's new national standards
The free desktop app for Windows, macOS and Linux opens, edits and saves files on your own computer.