Repair PDF

Files cut off at the end (interrupted download or copy), a missing or wrong internal index (xref), junk before the start of the PDF, a broken page list, references to parts that are gone and wrong internal lengths. In these cases the text, fonts and images stay exactly as in the original.

We read the file piece by piece, keep everything that is readable and write a new PDF with a clean structure. Before handing it over, we check that it opens, that no page was lost and that every page shows at least as much as in the original. As with our other tools, JavaScript, automatic actions and embedded attachments are removed.

Only when the structure can't be saved but the pages can still be drawn. Then we recreate each page from its image (up to 300 pages): it looks the same, but the text is no longer editable. Whenever possible the text goes into an invisible layer, so you can still search and copy it. The result always tells you when this happens.

No. The repair runs in your browser, in the background, and the repaired PDF is created in your device's memory.

Password-protected or restricted PDFs are not repaired here, so the protection is never removed by accident. Use Unlock PDF first (with the password, if it asks for one) and then repair the unlocked copy.

150 MB and up to 2000 pages per PDF. When the pages have to be recreated as images, the limit is 300 pages.

No. What is truly gone (for example, the part missing from an interrupted download) can't come back: we recover what is in the file. If nothing can be salvaged, we tell you instead of handing you an empty file.

No. It's free, with no sign-up and nothing to install.