Translate PDF
Translate a document's text into another language.
Your PDF is never uploaded. The text is extracted in your browser, shown to you, and only sent after you approve it.
What this does
This extracts a document's text in your browser, sends it to the AI provider you choose with your own key, and typesets the translation as a new PDF. What comes back is a clean document of translated text, not your original with the words swapped: German runs about a third longer than English and Indonesian shorter, so substituting in place would overflow some lines and leave holes in others. Paragraphs of prose survive that; images, tables, columns, headers, footers and the original pagination do not.
Use it when you need to read a document rather than reproduce it: supplier terms that arrived in a language you do not work in, a research paper you want to skim before paying to have it translated properly, an Indonesian regulation you need the substance of, or a manual whose instructions you have to follow today.
How it works
- Drop the PDF onto this page. Its text is extracted and shown to you, with an estimate of how much will be sent.
- Choose your provider and paste your API key. It stays in this browser and goes only to the provider it belongs to.
- Pick the target language. Thirteen are offered, and five of them come back as a text file rather than a PDF, for the reason below.
- Set the register under formality if it matters, and the paper size for the typeset result, under advanced options.
- Press Translate, then download the PDF, the plain text, or both.
The reason this produces a new document rather than an edited one is worth stating plainly, because plenty of tools imply otherwise. Text in a PDF is drawn at fixed coordinates; it does not sit in boxes that can grow. Replace an English sentence with its German translation and it is roughly a third too long for the space it has; replace it with Indonesian and there is a gap where the rest of the line used to be. Tools that do it anyway produce overlapping lines and clipped words. This one typesets the translation fresh, on the paper size you pick and with page numbers.
Five languages are delivered as text rather than as a PDF, and that is enforced rather than warned about. The font embedded in generated PDFs is Noto Sans, which covers Latin, Greek, Cyrillic and Vietnamese. Arabic, Chinese, Japanese, Hindi and Thai fall outside it, and a PDF built from glyphs a font does not have is a page of empty rectangles. So for those five you get a text file with a line at the top explaining why.
The text is cut into chunks of about 24,000 characters - smaller than the summariser uses, because a translation is roughly as long as its source and each request has to leave room for its own answer. Every chunk carries the same instruction: preserve meaning, structure and paragraph breaks, keep numbers, names and figures exactly, do not summarise and do not comment. Chunk boundaries fall between paragraphs, but a term rendered one way in one chunk can still come out differently in another.
Machine translation from a large model is good enough to work from and not good enough to sign. It handles ordinary prose well, technical vocabulary less reliably, and legal or contractual phrasing least reliably of all - which is to say it is weakest exactly where a translation matters most. For anything binding, this is a way to find out what a document says before engaging a translator, not a way to avoid engaging one.
What this tool cannot do
- The result is re-typeset from scratch. Images, tables, columns, headers, footers and the original page breaks are not reproduced.
- Arabic, Chinese, Japanese, Hindi and Thai are delivered as a text file, because the embedded font does not cover those scripts.
- The document's text is sent to the AI provider you choose, using your own key. The file itself is not.
- A scanned PDF has no text to translate; run OCR a PDF over it first.
Questions people ask
- Will the translation keep my original layout?
- No, and no tool working this way honestly can. The translation is typeset as a new document: paragraphs in order, on the paper size you choose, with page numbers. Images, tables, columns and the original page breaks are not reproduced. Translated text is a different length from its source, so putting it back into the original's fixed positions would overflow some lines and leave gaps in others.
- Does my file get uploaded?
- The file does not. The text extracted from it does, and only to the provider you named, posted by your browser with your own key in the header. There is no CekPDF server in that path. The panel shows you exactly what will be sent, and how much of it, before anything goes out.
- Why did Chinese come back as a text file instead of a PDF?
- Because the font embedded in generated PDFs is Noto Sans, which covers Latin, Greek, Cyrillic and Vietnamese but not Chinese, Japanese, Arabic, Hindi or Thai. A PDF built with a font that lacks the glyphs shows empty boxes where the words should be. Rather than hand you that, the tool delivers the translation as text with a note explaining it, ready to paste into a word processor that has the fonts.
- Is it good enough for an official document?
- Not for anything binding. A large language model translates ordinary prose well, technical vocabulary less reliably, and legal phrasing least reliably of all. Use it to learn what a document says, to decide whether it matters, and to brief a human translator on where to concentrate. Do not sign on the strength of it.
- What does a translation cost to run?
- Nothing here; your provider bills you. Translation costs more per page than summarising, because the output is about as long as the input rather than a fraction of it, and the text goes out in chunks of roughly 24,000 characters with one request each. The panel estimates the size before you send, and the character limit - 120,000 by default - caps what a single run can spend.