Repair a corrupted PDF
Attempt to recover a PDF that won't open or displays as corrupted. The file is rebuilt in your browser — first by repairing its internal structure, and if that's too damaged, by re-rendering each page from whatever can still be read.
Repair PDF
rebuild xref, or re-render pages as a fallback
Loading tool…
How to Repair PDF
- Add the damaged PDF. Drag the file onto the drop zone, even one that fails to open in other readers.
- Attempt repair. Click Attempt repair. The tool tries a structural fix first, and only falls back to a slower recovery method if that isn't possible.
- Check which recovery happened. The result tells you plainly whether the file was fixed with text intact, or recovered as page images because the damage was too severe.
- Download. Save the recovered PDF.
Two tiers of recovery, tried in order
Most PDFs that "won't open" have a genuinely fixable problem: a truncated or malformed cross-reference table, the internal index a reader uses to find each page and object. The first thing this tool tries is loading the file with a tolerant parser and immediately re-saving it — which rebuilds that index from whatever objects it could find. This fixes the majority of damaged PDFs completely, with the original text staying selectable and searchable.
If the damage goes deeper than that — the file is too corrupted to parse at all — the tool falls back to a second approach: rendering each page as an image using a different, more tolerant engine, then reassembling those images into a new PDF. This genuinely recovers whatever was visually on the page, but the result loses selectable text, since it's now an image rather than structured content. The tool tells you plainly which of the two happened, rather than presenting a degraded recovery as a full fix.
What this can't fix
Repair works on structural damage — a broken index, malformed objects, a truncated save. It can't recover data that was never written to the file in the first place, such as a PDF that was cut off mid-save with entire pages missing, or one where the underlying bytes are so scrambled that even a tolerant page-by-page render fails.
If neither recovery attempt succeeds, the file is genuinely unrecoverable with the data present in it — restoring from a backup or the original source is the only real option at that point.
Frequently asked questions
Will repair restore selectable text?
If the structural repair succeeds — which it does for most cases — yes, text stays fully selectable and searchable. Only the fallback image-based recovery, used when the file is too damaged for a structural fix, loses text selectability.
How do I know which recovery method was used?
The result clearly states which happened: a full structural fix with text intact, or a fallback recovery using page images. There's no ambiguity about which you received.
Is my file uploaded to a server?
No. Both recovery attempts run entirely in your browser.
What if neither recovery method works?
That means the underlying data needed to reconstruct the file is genuinely missing or too scrambled to read at all — not something any repair tool can recover. Restoring from a backup or the original source is the next step.
Why does my PDF say it's corrupted in the first place?
Common causes include an interrupted download, a crash during the original save, or a file that was only partially transferred. The structural fix here resolves the majority of these.
Can I repair a password-protected PDF?
Remove the password first with the Remove Password tool, then run the unlocked file through repair.