Translate a PDF into 26 languages — and a scanned one keeps its pages, photographs and slide layout, with the new wording painted where the old was. Free, no sign-up, no watermark.
Drop your PDF files here
or click to browse — Select multiple files for batch conversion
☁️ Only the wording is sent — your PDF is opened and read in this browser tab. The text travels to our translation service and back, and the translated text is laid out as pages on our server; the file itself is never uploaded, and nothing is stored on either leg. Why FileHugger
Every other PDF translator puts this behind a paywall, a page cap or an account. It is the one tool the big converter sites all agree to charge for. Here it is free, because the whole site is: the document is read in your browser, only the wording is sent to be translated, and the pages are laid out and handed back. The structure survives — headings stay headings and lists stay lists — and the result is a real PDF you can print or send on, not a wall of text.
PDF is the universal document format: it locks layout, fonts and images so a file looks identical on every device and printer. Scans, invoices, forms, e-books and reports all travel as PDF. Because a PDF is a container rather than an image, turning its pages into JPG or PNG pictures — or bundling pictures into a PDF — are among the most common file tasks there are.
Look at what is free on the big converter sites and what is not. Merging, splitting, compressing, converting to Word — all free, everywhere. Translating a PDF is the exception: iLovePDF, Smallpdf and the rest put it behind a paid tier, a page cap or at least an account. It is the most consistently paywalled tool in the category.
There is a reason for that, and it is worth being open about: translation costs money to run, unlike the tools that work entirely in your browser. We do it free anyway, in twenty-six languages, with no account, no watermark and no per-page charge.
Most cheap translators hand you a text file, which is a poor trade for a document you needed to send to somebody. The pages are read for their structure first — which lines are headings, which belong to the same paragraph, which are list items — so the translation goes back into a document with the same shape, laid out properly and paginated as a real PDF.
This is where a lot of PDF translators quietly fail. The handful of fonts a PDF is guaranteed to contain cover Western European characters only, so a tool that writes the PDF directly cannot produce Greek, Cyrillic, Arabic, Chinese, Japanese, Korean — or even Turkish, where ı, ğ and ş fall outside the set. Translating into those languages and then dropping half the letters is worse than useless. The pages here are laid out by a full browser engine, which has no such limit.
This is machine translation, and being free means it runs on an open engine rather than a commercial one. That shows in a specific, predictable way: full sentences come out well — a contract clause or a dosage instruction reads correctly — while short fragments on their own do not. A heading of one or two words has too little context to translate from, so “Overview” can come back as something shorter or blunter than it should be. Expect the body of a document to be good and its headings to be rough.
Figures are the part that matters most, and they are checked rather than trusted. These engines sometimes corrupt numbers in a short or number-heavy line — turning a year into a longer number, or merging a percentage and an amount into one figure. Every block’s digits are compared with the digits that went into it, and any block whose numbers changed keeps its original wording instead. The result tells you how many blocks that happened to. An untranslated line is obvious and recoverable; a wrong figure in a financial document is neither.
The same principle catches the other way these engines fail. A fragment with too little context to work from sends them into a repetition loop: a spreadsheet column headed Engineer came back, in testing, as the same word three times over. That is worse than the English word left alone, because it looks like the tool is broken rather than like the tool declined. So a translation that repeats a word the original did not, or that comes back three times the length of a one- or two-word original, is refused and the original kept. It costs the occasional correct translation — some single words genuinely need three in another language — and that is the trade being made deliberately.
None of which makes it a translator. It is a genuine help with a document in a language you do not read, and it is not something to put a contract, a diagnosis or a legal filing in front of anyone on the strength of.
The layout is not reproduced either, and that is a deliberate choice rather than a shortcoming: translated text is a different length from the original, so forcing it back into the same boxes produces overlapping, clipped text. Words, headings and lists come across; fonts, columns, images and exact positioning do not.
This is worth knowing before you use this tool at all. A PDF has to be rebuilt to be translated, because its text is fixed to coordinates. A Word document, an Excel workbook, a PowerPoint deck and an EPUB do not: each is an archive of XML, so the wording can be replaced inside the file and the fonts, tables, images, headers and page setup come back untouched. If the PDF you are holding started life as one of those and you can still get at the original, translating it there gives a visibly better result — and hands you back an editable file rather than a fresh PDF.
There is a size limit of roughly forty to sixty pages per document, so that one request cannot occupy the service for everybody else. And a scanned PDF has no text in it to translate — run OCR PDF on it first, then come back.
The PDF itself never leaves your browser. It is opened and read here; only the extracted wording is sent to be translated, and only the translated wording comes back to be laid out. Nothing is stored on either leg. That is a weaker guarantee than the tools that never send anything at all, which is why it is stated plainly rather than buried — and it is the reason this tool, unlike most of the site, has a server behind it.
Yes — there is no sign-up, payment or watermark. Server-processed tools use the published file-size, duration and fair-use limits so jobs remain available to everyone; those are technical safeguards, not a paid tier.
Yes. Drop in as many as you like and each is translated in sequence with its own progress status, then downloaded individually or all together as a ZIP. Long files are sent for translation in batches, so a stack of large documents can reach the fair-use ceiling — if it does, the message says so and waiting a few minutes is all it takes.
Yes — this tool needs a full office engine, so your file is sent to our server over an encrypted (HTTPS) connection, converted there, and deleted immediately after the PDF is delivered. It is never read for any other purpose, never shared, and never stored. Files up to 50 MB are accepted.
It is free and there is no account, no watermark and no page cap beyond a size limit of roughly forty to sixty pages per document, which exists so one request cannot occupy the service for everybody. The catch is quality, and it is a specific one: free translation runs on an open engine, which handles full sentences well and one- or two-word headings poorly. Figures are checked rather than trusted — any block whose numbers come back different from the numbers that went in keeps its original wording instead of being published wrong, and so does any block that came back repeating itself, which is how these engines fail on a heading with too little context to work from. It is a good draft and not a substitute for a translator on anything legal, medical or contractual.
filehugger.com