My Tool Studio
PDF

Repair PDF

Repair PDF tries to fix a PDF that won't open, shows blank pages or throws a file is damaged error. It checks the file first and tells you what is wrong: a download that was cut off, junk bytes before the header, a broken cross-reference table, a lost page list, or a file that isn't a PDF at all. Then it runs up to four recovery methods, including the qpdf repair engine, and checks that every recovered page really opens. You get the version that saves the most pages, and you can pick another if you prefer. Several files can be repaired at once, and none of them are uploaded.

Always freeNo sign upRuns in your browser

How to use

01

Add the damaged files

Drop one or more files of up to 200 MB each. Any file is accepted, because a damaged PDF sometimes turns out to be another format with the wrong name.

02

Set the options and repair

Enter the password if the file has one, and choose when image rescue should run. Click Repair PDF. Each method runs in turn and every result is checked page by page.

03

Pick a result and download

Read what was found, compare the methods in the table and keep the recommended one or choose another. Preview the pages, then download the repaired PDF, or all files as a ZIP.

Why Repair PDF

What the checks look for

Before any repair runs, the file is scanned byte by byte. Each finding points to a different cause, and it tells you whether a repair can fully succeed.

FindingWhat it usually meansCan it be fixed?
End-of-file marker missingDownload or copy was cut offPartly: the pages that arrived
Stray bytes before %PDFExtra data added by an email or upload systemYes, fully
Extra bytes after %%EOFPadding or leftovers from a failed saveYes, fully
Broken cross-reference tableThe file's index points to the wrong placesUsually fully
Page list damagedPages exist but are no longer linkedUsually, order may need checking
Not a PDF at allAnother format with a .pdf nameRenamed or converted

How the recommended result is chosen

Every method's output is opened again and each page is parsed. The method that gives the most readable pages wins. If two methods tie, real pages beat images, and qpdf beats the others because it keeps the most of the original structure.

You can still pick another row in the results table. That helps when, say, the page tree rebuild found an extra page the others missed but put it in a different order.

Common questions

Why won't my PDF open?
The usual causes are a download or email attachment that was cut off, a file copied from a failing USB stick, an app that crashed while saving, or a web page saved with a .pdf name. Repair PDF checks for each of these and lists what it found at the top of the result.
Can every damaged PDF be repaired?
No. Repair can rebuild the structure of a file, but it can't bring back data that is missing. If a download stopped halfway, the pages that were never downloaded are gone, and the tool recovers the ones that arrived. For a complete copy, download the file again from its source.
What do the four PDF repair methods do?
qpdf rebuilds the file's index of objects and rewrites it. pdf-lib reads every object one by one and copies the pages into a fresh file. Page tree rebuild collects page objects that lost their place in the page list. Image rescue saves each page that can still be drawn as a picture.
What is image rescue when repairing a PDF?
It is the last resort. Pages that the PDF viewer can still draw are saved as images at 150 DPI in a new PDF. The pages look right, but their text can no longer be selected or searched. By default it runs only when the other methods lose pages. Run the result through OCR PDF to make it searchable again.
My damaged PDF turned out to be a Word file. How did that happen?
Files are sometimes saved or renamed with the wrong extension, and a .docx, an image or a web page with a .pdf name won't open in a PDF reader. The tool recognises common formats and offers the file back with the right extension. JPG and PNG images are also turned into a one-page PDF.
Can I repair a password-protected PDF?
Yes, if you know the password. Type it in the password box before you click Repair. The qpdf method repairs the file and keeps the same password on the result, and image rescue can also use it. Without the password an encrypted file can't be read, so it can't be repaired.
Will repairing a PDF change its content or quality?
The qpdf, pdf-lib and page tree methods copy the real pages, so text, fonts and images stay exactly as they were. The qpdf result also keeps bookmarks and form fields, while pdf-lib and page tree results keep the pages only. Image rescue is the only method that turns pages into pictures.
Is it safe to repair confidential PDFs with this tool?
Yes. The files are read and rewritten by code running in your browser tab, including the qpdf engine, which runs as WebAssembly. Nothing is uploaded, so bank statements and legal papers stay on your device.
Why does the repaired PDF have fewer pages than the original?
Some pages could not be rebuilt, usually because their data is missing or badly damaged. The results table shows how many pages each method recovered, and the expected count comes from the file itself. Try Image rescue set to Always, which sometimes saves a page the other methods drop.
How can I stop PDFs from getting corrupted?
Let downloads finish before opening them, eject USB drives properly, and avoid editing a PDF that is still syncing to a cloud folder. Keep a second copy of important files. When a download fails, a fresh download is almost always better than any repair.

More PDF tools

View all

Merge PDF

Combine PDFs and images into one PDF

PDF

Split PDF

Split a PDF by ranges, size or bookmarks

PDF

Compress PDF

Reduce PDF size to a target or by level

PDF

PDF to Excel Converter

Extract PDF tables to Excel or CSV

PDF