PDFLabs
PDFLabs › User guide
The user guide explains how to merge, split, compress, and convert PDFs in this browser.
Merge PDF joins files in the order you add them. You can change that order in the list, and files are not sent to a server.
Pages are joined from the top of the list. You can change the order before merging. One PDF is not enough, so add another file.
The merged PDF is saved on this device. PDFLabs processes files in the browser and does not send or store them on a server.
After merging, Split PDF can save every page, each range, or only the pages you pick.
Merge PDF takes files you drop or choose. You can add up to 40 files and 120MB at once, and you can drag the list to change the order.
A password-protected PDF has to be unlocked before merging. Enter the password you already know in Remove a PDF password. The password is not guessed.
After merging, Arrange pages can rotate, duplicate, delete, and reorder pages from the preview. You cannot delete every page.
Split PDF saves one PDF by page, by range, in bundles, or as only the pages you pick. Files are not sent to a server.
Split PDF uses one of four ways: every page, ranges, bundles, or selected pages. Every page saves each page on its own. Ranges start a new file at each comma, for example 1-3, 5, 8-10. Bundles set how many pages go in one file, from 1 to 100. Selected pages gather the pages you pick into one PDF.
One result arrives as a PDF. Several results arrive as a zip file.
The files do not stay on a server. The split runs in the browser, and the result is saved on this device.
Split PDF takes one PDF. Drop it or choose it, and one batch can be up to 120MB.
A password-protected PDF is unlocked with the password you already know before splitting. The password can use only letters, digits, and symbols, and a forgotten password cannot be removed.
Deleting pages from the preview uses Arrange pages. Split PDF saves by page number and range, and it does not pick pages off the preview to delete them.
Compress PDF redraws pages as images to reduce the size. The text cannot be selected with a mouse, and if the result is larger than the original, keep the original.
High quality keeps more detail. The size change may be small. Recommended is sized for reading on screen. Smallest is the smallest of those three, and text may look soft. Gray redraws pages as gray images at the medium quality. Custom sets the resolution and JPEG quality. Gray removes the color.
Text in the compressed PDF cannot be selected with a mouse, because the pages are redrawn as images.
The compressed file is saved on this device. It is not sent to a server or stored there.
Compress PDF can reduce several PDFs at once. One result arrives as a PDF, and several results arrive as a zip file.
If the result is larger than the original, you keep the original. A file that is already small may not get any smaller.
Selecting the text again takes OCR, which reads the printed text. Compress redraws each page as an image.
Convert PDF turns images and documents into PDFs, or sends a PDF to images, Word, Excel, and PowerPoint. Text and table content carry over. Formatting from the original editor does not.
JPG, JPEG, PNG, WEBP, GIF, and BMP go into one PDF in the order you choose. They fit the paper size, and a GIF uses only the first frame.
DOCX, XLSX, XLS, PPTX, TXT, CSV, MD, JSON, HTML, and EPUB become PDFs. Headings, paragraphs, tables, and DOCX images carry over. Complex formatting from the original app and PPTX animations are not kept. Sheets and CSV include the first 2,000 rows and 20 columns.
A PDF can be saved as Word, Excel, PowerPoint, HTML, EPUB, SVG, or Markdown. Only text and table content carry over, and formatting from the original editor does not stay. Pages can also be saved as PNG, JPG, or WEBP, and several pages arrive as a zip file. Text that is in the PDF can be saved as TXT. Text that exists only as an image is not extracted.
Web page to PDF takes an http or https address. If the browser blocks that address, the page is not fetched, and scripts and remote images are left out.
Image to PDF takes JPG, JPEG, PNG, WEBP, GIF, and BMP. Document to PDF takes DOCX, XLSX, XLS, PPTX, TXT, CSV, MD, JSON, HTML, and EPUB, and several documents arrive as a zip file.
A scanned PDF needs OCR to read the printed text before it can become Word. If the text exists only as an image, the paragraphs are not carried over.
Reading a scanned PDF reads printed text. Handwriting is often missed, and after you choose a language the result can be a text file or a searchable PDF.
It reads printed text in a scanned PDF. Handwriting is often missed. It also reads printed text in an image.
After you choose a language, the result can be a text file or a searchable PDF.
The PDF contents do not go to a server. Recognition runs on this device, and only the recognition program is downloaded.
Reading a scanned PDF reads one of Korean, English, Japanese, simplified Chinese, traditional Chinese, Spanish, Portuguese, French, German, and Indonesian.
A tilted page is straightened when auto rotate is on. Auto rotate starts on.
From a card or a receipt it picks values that look like an email, a phone number, an amount, or a date. Handwriting is often missed.
Arrange pages rotates, duplicates, deletes, and reorders pages from the preview. Files are not sent to a server.
Select pages in the preview, then use Delete selected. You cannot delete every page, and at least one page has to stay.
Select pages, then rotate left or right by 90 degrees. Save writes that direction into the file.
Drag a page onto another page to change the order. Reverse order flips the whole document.
Copy selected places one more copy right after the page. Nothing is copied when no page is selected.
The organized file is saved on this device. PDFLabs processes the file in the browser and does not send or store it on a server.
A PDF password is set or removed only with a password you already know. It is not guessed, and the file is not sent to a server.
Set a PDF password by entering the password used to open the file. The password for opening and the password for changing can be different.
Printing starts allowed, and copying and editing start off. Some programs can ignore these marks.
The password can use only letters, digits, and symbols. A forgotten password cannot be removed.
Remove a PDF password by entering the password you already know. If the file cannot be opened that way, the pages are rebuilt as images, and then the text cannot be selected with a mouse.
The unlocked file is saved on this device. The password and the PDF are not stored on a server.
Add a watermark places faint slanted text in the center of the page. You can set the words and how faint they are, and the file is not sent to a server.
The watermark goes across the center of the page. The angle starts at 35 degrees, and the words, size, color, and opacity can be changed.
The watermark is drawn on top of the page. It does not edit the original paragraphs.
Add a signature places words or an image on the page. The time on this device is written with it, and it is not a certified electronic signature.
Add page numbers places the number at the top, bottom, bottom right, or bottom left. The watermark stays slanted in the center.
The finished PDF is saved on this device. It is not sent to a server or stored there.
Add page numbers places a number at the bottom or top of the page. {n} is the number and {total} is the page count, and the file is not sent to a server.
Page numbers go at the top, bottom, bottom right, or bottom left. {n} is the number and {total} is the page count.
The first number can be any number you choose. It starts at 1.
Add a header and footer places the same words at the top and bottom. Add page numbers places the number.
The numbers are drawn on top of the page. They do not edit the original paragraphs.
The numbered PDF is saved on this device. It is not sent to a server or stored there.
Repair a PDF saves the file again to tidy its structure. If it still will not open, the pages are rebuilt as images, and not every kind of damage can be fixed.
Repair a PDF saves a PDF that can be opened, so the structure is tidied. The file is not sent to a server.
A file that will not open is rebuilt with its pages as images. The text then cannot be selected with a mouse.
Not every kind of damage is fixed. A file can still fail to open after it is saved again or rebuilt as images.
Repair a PDF does not guess the password of a protected file. A password you already know is removed in Remove a PDF password.
The repaired file is saved on this device. It is not kept on a server.
Build a contents list treats large text lines as headings and turns them into bookmarks. They may differ from the contents the document states.
Text size is measured on every page, and lines that are large for that document are treated as headings. Their text and page number become bookmarks.
Cover pages and large advertising text also look like headings, and small subheadings are left out. That is why the result can differ.
They appear in the contents panel of the program that opens the PDF. The original pages are not changed, and only jump targets are added.
Building it again adds new bookmarks on top. Existing bookmarks are not deleted.
Text in a scanned PDF is a picture, so its size cannot be measured. Read the text with OCR first.
Compare PDFs lines up the text lines of two PDFs and shows lines that are missing or changed. Differences in layout or pictures are not checked.
The text lines taken from both files are lined up in order, and lines that appear on only one side and lines whose text changed are marked.
Pictures, table column widths, and margins are not compared. A file where only a photo changed comes out the same.
When paragraphs change order, they come out as removed and added lines. Moved text is not labelled as moved.
Text in a scanned PDF is a picture, so there are no text lines to compare. Read the text with OCR first.
The result is shown on this screen and is not saved as a file. Neither PDF is sent to a server.
Extract from a PDF pulls images, attachments, font names, links, notes, bookmarks, and tables out of one PDF. Only the kind you choose is saved.
Images, attachments, font names, links, notes, bookmarks, and tables. Tables are guessed from the spacing of text lines and saved as CSV or XLSX.
Fonts are exported as names only. The font files themselves are not exported.
A PDF has no table structure, so columns are guessed from the spacing of text lines. Merged or empty cells can shift.
Images come out as they are stored in the PDF. When there are several, they are packed into one file.
On this device. Neither the PDF nor the result is sent to a server.
Web page to PDF turns a page into a PDF when you enter an http or https address. If the browser blocks that address, the page cannot be fetched.
An address that starts with http or https. Pages that need a sign-in may not open.
Scripts and remote images are dropped, so the result can differ from what you saw on screen.
When the browser blocks that address. Sites that block being read from elsewhere fall into this case.
This browser opens the address itself. PDFLabs does not open it on your behalf.
On this device. It is not stored on a server.
Edit PDF details changes the title, author, subject, and keywords, or clears them all at once. Personal details in the body stay as they are.
The title, author, subject, and keywords stored in the file information, and the creator entry is cleared.
No. Only the file information is cleared and the text on the pages stays. To cover the body, use Mask sensitive text.
When you send a file on, an author name or the program that made it is not revealed. What is written in the body stays.
Turning on Remove details empties the title, author, subject, keywords, and creator at once.
On this device. The file is not sent to a server.