OCR PDF
Runs on our server, not on your device.
OCR PDF online: make a scanned PDF searchable, with text you can select and copy.
- Drop in a scanned PDF and see how many pages need reading
- Press Make searchable; each scanned page is read in turn
- Download the searchable PDF, or just the text as a .txt file
A free account includes 5 AI credits a month, Pro 100.
Your PDF stays on this device. A picture of each scanned page is sent to an AI model to be read, and is not kept.
Every page in this PDF already has text, so it is searchable as it is. There is nothing to read.
Check the result before you rely on it. Reading and converting documents is never perfect, and a misread number is easy to miss.
Save this tool
Keep OCR PDF handy
Use your browser's Share menu, then “Add to Home Screen”.
About this tool
A scanned PDF is a stack of photographs. It looks like a document, but there is no text in it at all, so you cannot search it, select a sentence or copy a figure out of it. OCR PDF reads each scanned page and lays the words it finds over the page as invisible text, which is how a searchable PDF works: the page looks exactly as it did, and search, selection and copying now find the words under the picture. You also get everything it read as a plain text file. Pages that already have text are left alone, so a document that is half typed and half scanned only has its scanned pages read, and the tool says how many it skipped. The reading is done by an AI vision model, which copes far better than older OCR engines with crooked scans, phone photos of pages, faint print and tables. It is still a reading, not a guarantee: it can misread a smudged figure, and handwriting is hit and miss, so check anything that matters against the page. Unlike most tools here this one cannot run entirely on your device. Your PDF stays in your browser, but a picture of each scanned page is sent through our server to the AI model that reads it, and nothing is kept once the page has been read. Each scanned page costs one AI credit, and the count is shown before you start. A free account includes 5 credits a month, Pro 100.
Frequently asked questions
What does OCR PDF actually change in my file?
It adds an invisible text layer over each scanned page, the standard way a searchable PDF is made. The pictures are not touched, re-compressed or redrawn, so the pages look exactly as they did. Search, selecting text and copying now find the words that were read. Pages that already had text are left exactly as they were.
Is my PDF uploaded?
The PDF itself stays in your browser: finding the scanned pages, turning them into pictures and writing the text layer all happen on your device. What leaves it is one picture of each scanned page, sent through our server to the AI model that reads it, and the lines read come back. Nothing is stored once a page has been read. Pages with text of their own are never sent.
How accurate is it?
Usually very good on printed text, including crooked scans, phone photos of pages and faint print, which are where older OCR engines struggle. It is still a reading rather than a copy: a smudged number can come out wrong, and handwriting is hit and miss. Check figures that matter against the page. The invisible text is placed line by line, so selecting it follows the lines on the page closely, though not always to the exact letter.
Why does it cost credits?
Because every scanned page is read by an AI model, which costs real money each time. One page is one credit, and the tool counts the pages that need reading before you start, so you know the price first. A free account includes 5 credits a month once its email is confirmed, and Pro includes 100. Pages that already have text cost nothing.
Which languages does it read?
Any language the model can read, which covers the common European languages and many more. You can type the language as a hint for a poor scan, but it is not required. The text file holds everything read in any script; the invisible layer in the PDF is drawn in a font covering Latin, Greek and Cyrillic, so search and copy in the PDF work best in those.
Can I just get the text out?
Yes. When you press Download you can choose the searchable PDF or a plain text file with the words read, page by page, and Copy puts the text on your clipboard. If you only have a photo rather than a PDF, the Image to Text tool reads that.
Help improve this tool
Found a bug, want a change, or need a tool we don't have?
You have an unsent draft.
Thanks — that's with us. You'll find the reply in your inbox.Go to my messages