PdfLuxe › Guides › Best PDF OCR
Best PDF OCR — make scanned PDFs searchable, free
Best PDF OCR, OCR PDF online free, make a PDF searchable — searches that start with a scanned document that behaves like a photograph: you can look at it, but you cannot search it, copy from it, or convert its words. This guide explains what OCR actually does, how PdfLuxe OCR PDF turns scanned pages into searchable text in your browser, and the limits stated plainly — one run reads up to 10 pages, print reads far better than handwriting, and accuracy is decided mostly by the scan itself, long before any tool sees it.
Updated 19 November 2026 · Matches PdfLuxe v345.
By the PdfLuxe team · Contact
The short answer
The best PDF OCR tool does three things well: it reads the page accurately and tells you how accurate the read was, it gives you back a document that behaves like a real document — searchable, selectable, copyable — and it does not require you to upload private scans to a stranger's server to get there. PdfLuxe OCR PDF is built to that standard. It reads scanned pages on your device with the Tesseract engine, in seven languages, shows a confidence percentage for every page, and returns the words as a .txt file or as a searchable PDF whose pages look identical but carry the text as an invisible layer. It is free, with no sign-up and no watermark. Its honest limit is stated here rather than discovered later: one run reads up to 10 pages, so a long scan is done in parts — the section below explains how, and why the limit exists.
How to OCR a PDF online for free, step by step
- Open OCR PDF on PdfLuxe and add your scanned PDF — or pick a JPG or PNG photo directly in the read-a-photo field.
- Choose the language printed on the page — English is pre-selected, and Lithuanian, Hindi, Russian, German, Spanish and French are also available — and tick the deskew option if the scan sits at an angle.
- Click Run OCR. On the first run the engine and language pack (a few MB) download once; each page is then read on your device, up to 10 pages per run, and the recognised words appear with a confidence percentage per page.
- Download the words as a .txt file, or download the searchable PDF — your original pages with the recognised text added as an invisible layer, so the file looks unchanged but can be searched, selected and copied.
What OCR actually does to a scanned PDF
A scanned page looks like text to you, but to a computer it is only a photograph: thousands of pixels, no letters, nothing to search, select or copy. OCR — optical character recognition — is the process of reading that photograph and working out which letter each shape is. OCR PDF does the reading with the Tesseract engine running inside your browser, and it reads each page twice, with two different layout settings, keeping the clearer read — which is why the result arrives with a confidence percentage per page rather than a silent guess.
The two outputs serve different needs. The plain text is what you want for copying into a document, a form or a translation tool. The searchable PDF is subtler and often more valuable: it keeps your original page images exactly as they are and lays the recognised text invisibly behind them, aligned to the page. The document looks identical — same scan, same stamps, same stains — but Find now works, text can be selected and copied, and the other converters can read it too: once a scan is searchable, PDF to Word, PDF to Excel and PDF to Text all work on the words OCR added. A box of scanned archives becomes a library you can search in seconds.
The 10-page limit, stated plainly
One OCR run reads up to 10 pages. This guide puts that in its own section because it is the fact about this tool you most need before you start a 200-page scan, and because the reason is structural rather than commercial: the reading happens on your device, in a browser tab, and a run is kept to a size that tab handles well on ordinary hardware. There is no paid tier that raises it — PdfLuxe has no paid tier at all.
Working with a longer document is a workflow, not a wall. Split the scan into chunks with Split PDF — every 10 pages is one tap with the Every N pages helper — and run OCR on each part, then merge the searchable parts back together with Merge PDF if you need one file at the end. Or simply run the tool again on the next pages of the same document. Either way, the whole book gets read; it just gets read in sittings, the way a person would read it too.
Getting the best read: language, resolution and a straight page
OCR accuracy is decided mostly before the tool ever sees the page. The right language matters more than people expect. Seven languages are available — English, Lithuanian, Hindi, Russian, German, Spanish and French — and picking the page's language tells the reader which alphabet and dictionary to expect, which is decisive for letters with marks and accents: Lithuanian ą, č, ę or German ä, ö, ü are only recognised reliably when the reader is expecting them. Resolution is the second lever: scan at 300 dpi if you can, because letters need pixels to be recognised — a small, soft image gives the reader shapes, not characters. Straight and flat is the third: photograph pages from directly above, evenly lit, and tick the deskew option for a tilted scan; it corrects an angle, but it cannot undo a curling page or a shadow across the text. Clean contrast is the fourth: black print on white paper reads best, and faded, stained or reversed-out text will produce more errors wherever it appears. Check the confidence percentage per page, and proofread names, dates and numbers in anything that matters.
What OCR cannot do, honestly
Handwriting is the big one: this reader is built for printed text, and cursive or handwritten notes come back fragmentary at best — no setting changes that, and a guide that promised otherwise would be selling you proofreading work disguised as a result. Very small print, decorative display type, and badly blurred or shadowed scans also read poorly, for the same reason: the engine reads shapes, and damaged shapes give damaged words. And OCR adds a text layer; it does not restore a document. A searchable PDF of a bad scan is a bad scan you can search — which is still, for most archives, exactly the improvement that matters.
How other services meter the same job
OCR is one of the jobs freemium services most often keep behind the paid tier, and the published figures show it. On iLovePDF, advanced OCR PDF is a Premium feature, and its per-task table gives free OCR 1 file per task with a 15 MB file-size cap. On Sejda, OCR sits inside the free service's published terms — documents up to 200 pages or 50 MB and 3 tasks per hour — and paid plans raise OCR to 100 pages. On Smallpdf, OCR is listed among the Pro features on its pricing page, and free tools carry the daily download limit the page marks without publishing a number. On PdfLuxe, OCR is a free tool whose limit is a page count per run — 10 pages, stated above — rather than a paywall: run it as many times as the document needs.
Competitor figures are from the providers' own pricing and tool pages, checked 14–15 November 2026: iLovePDF's Premium feature list and per-task table, Sejda's upgrade page and free-service sentence, and Smallpdf's pricing page, which lists OCR among Pro features and marks a daily download limit without publishing the number.
Privacy: your scans stay on your device
Scans are often the most private documents a person owns — ID papers, medical letters, bank statements, signed forms. OCR PDF's default engine reads them in your browser: the pages never leave your device, and there is no server copy to worry about. The first run downloads the OCR engine and the language pack (a few MB), so it needs an internet connection once; after that, the reading itself happens on your device. The tool also offers optional cloud OCR engines that run on your own key from AI Settings; those do send each page image to the provider you chose, under your key, and they are labelled as cloud options inside the tool so the choice is always yours and always visible.
Try it on the scan you cannot search
Pick the scanned document that frustrated you last — the one where Find found nothing. Run OCR on it, download the searchable PDF, and search it for a word you know is on page two. That single test, on your own document, tells you more about OCR than any guide, including this one.
Questions & answers
What is the best PDF OCR tool?
The best PDF OCR tool turns a scanned page into text you can search, select and copy, tells you how much to trust each page, and does not make you upload private scans to get it. PdfLuxe OCR PDF reads pages on your device with the Tesseract engine, shows a confidence percentage per page, and returns either the plain text or a searchable PDF with the text as an invisible layer — free, with no sign-up and no watermark. Its stated limit is 10 pages per run, which this guide covers honestly.
How do I make a scanned PDF searchable for free?
Open OCR PDF on PdfLuxe, add your scanned PDF, choose the language printed on the page, and click Run OCR. On the first run the engine and language pack download once; the pages are then read on your device. Download the searchable PDF: your original pages, unchanged to look at, with the recognised text added as an invisible layer behind them — so Find works, text can be selected and copied, and other converters can read the words.
How many pages can PdfLuxe OCR read at once?
Up to 10 pages per run. That limit is stated plainly because it is real: the reading happens on your device, and a run is kept to a size a browser tab handles well. For a longer document, split it into chunks with Split PDF and run OCR on each part, or run the tool again on the next pages. Each page is read twice with two layout settings and the clearer read is kept, with a confidence percentage shown per page.
Which languages can OCR read?
English, Lithuanian, Hindi, Russian, German, Spanish and French. Choose the language printed on the page before running — it tells the engine which alphabet and dictionary to expect, and it makes a real difference for letters with accents and special marks. The language pack for your choice downloads on the first run, a few MB, and is cached afterwards, so later runs start straight away.
What is the difference between the text download and the searchable PDF?
The .txt download is just the recognised words — best for copying into a document, a form or a translation tool. The searchable PDF keeps your original scanned pages exactly as they look — same scan, same stamps and marks — and adds the recognised text as an invisible layer behind them, so the file looks unchanged but can be searched, selected and copied, and read by search tools and screen readers. A scan archive converted this way becomes a library you can actually search.
Why is some of the recognised text wrong?
OCR reads shapes, so anything that blurs the shapes costs accuracy: low-resolution scans, tilted or shadowed pages, faded print, or the wrong language selected. A straight, clean 300 dpi scan with the correct language set reads far better than a casual photo. The confidence percentage beside each page tells you how much to trust it — proofread names, dates and numbers in anything that matters, because those are exactly where a single wrong character does the damage.
Can OCR read handwriting?
Not reliably. The engine is built for printed text — books, letters, forms, receipts — and handwriting, especially cursive, comes back fragmentary at best; no setting changes that. Very small print and decorative display type also read poorly. If the document matters and it is handwritten, transcription by a person is still the honest route, and a tool that tells you so is worth more than one that returns confident nonsense.
Are my scanned pages uploaded for OCR?
Not with the default engine. OCR PDF's on-device Tesseract engine reads pages in your browser, and your pages never leave your device; only the first run needs the internet, to download the engine and language pack. The tool also offers optional cloud OCR engines that work with your own key from AI Settings — only if you choose one of those are page images sent to that provider, under your key, and the tool labels them as cloud options.
Related tools
This guide is one of 92 step-by-step guides on PdfLuxe — 151 free tools, no sign-up, no watermark. Questions? Email support@pdfluxe.com.