Duplicate File Detector

How to use the duplicate file detector

1. Choose the files to compare — a whole folder's worth is fine.
2. Select Find duplicates.
3. Review each set and delete the copies you do not need.

About this tool

Compares files by content and reports which are byte-for-byte identical, together with how much space keeping one copy of each would free.

Matching is on a SHA-256 hash of the contents, not on the name or the size, and that distinction is the point. Two files with different names are still duplicates — `invoice.pdf` and `invoice (1).pdf` usually are. Two files of the same size usually are not, since size collides constantly and tells you nothing about content. A name-based check finds the obvious cases and misses the ones that actually accumulate.

A hash match means identical content, for any practical purpose. Two different files sharing a SHA-256 has never been demonstrated, and the odds are far below the odds of the disk misreading them anyway.

What this does not do is find files that are merely similar. Two exports of the same photograph at different quality settings look alike and hash completely differently, because one byte of difference changes the entire hash. That is what makes the result trustworthy rather than approximate.

Every file is read in the page and hashed there, so nothing is uploaded. Nothing is deleted either: the result is a report, and which copy to keep is left to you, since the tool has no way of knowing which one something else on your machine is pointing at.

Common uses

• Clearing duplicate photos out of a downloads folder.
• Checking whether a backup copy is still identical to the original.
• Confirming two downloads of the same file match.

Frequently asked questions

Does it delete anything?

No. It reports what is duplicated and leaves the deleting to you — a page cannot remove files from your disk, and which copy to keep depends on what else points at it.

Will it find similar photos?

No, only identical ones. Two versions of the same photo saved at different quality differ by bytes and hash completely differently. Matching content exactly is what makes the result reliable.

Can two different files show as duplicates?

In principle a SHA-256 collision is possible; in practice none has ever been found, and it is far likelier that a disk misread the file.