Guide

How to find duplicate files on Mac

Duplicates are the easiest gigabytes to give up, because deleting one is not a judgement call — an identical copy stays behind. The hard part is proving two files are identical rather than similarly named.

Where the copies come from

Nobody duplicates files on purpose. They arrive through AirDrop and Messages saving a second copy, downloads landing twice as file, file-1 and file-2, camera imports run from more than one machine, a folder dragged instead of moved, and old project folders copied "just in case" before a rewrite. On a Mac that has been in use for a few years, tens of gigabytes of exact copies is normal.

Name matching is not enough

Searching by name finds report-final-2.pdf next to report-final.pdf and calls them duplicates, which they may not be. It also misses the case that matters most: the same file saved under two different names, which no name search will ever pair up. The only reliable test is the contents. Compare size first as a cheap filter, then compare the bytes of the candidates — two files are duplicates when every byte matches, whatever they are called and wherever they sit.

What macOS gives you, and where it stops

  • Photos has a Duplicates album on recent macOS versions. It works well inside your photo library and does nothing outside it.
  • A Finder Smart Folder can filter by name, kind and size, which narrows a search but never proves two files are identical.
  • In Terminal, shasum can hash a shortlist of files you already suspect. Hashing an entire disk this way is slow and easy to get wrong.
  • Nothing built into macOS scans the whole volume for exact copies and groups them for you.

Rules that keep the delete safe

01
Always keep one copy of every file A duplicate finder should never be able to empty a group. If a tool can leave you with zero copies, it is the wrong tool.
02
Decide which copy is the keeper before you start The filed copy in its proper folder, the newest, or the oldest — pick a rule and apply it consistently instead of choosing file by file at two in the morning.
03
Leave library packages structurally alone Photos and Music libraries can be compared, but copies inside them should be resolved by the app that owns the library rather than by pulling files out of the package.
04
Send copies to the Trash, not to oblivion Reversibility costs nothing until you empty the Trash. Keep the decision undoable for a day.

How Sift does it

Sift matches byte for byte rather than by name, so a renamed copy is still caught and two unrelated files that happen to share a name never are. Copies are grouped, one copy of every file is always kept, and the keep rule is yours: the filed copy, the newest or the oldest. Photos and camera imports are compared, never rearranged. The scan that finds them is free; moving copies to the Trash is part of the yearly plan.

Find your duplicates with Sift

Common questions

Does macOS have a built-in duplicate file finder?

Not for the whole disk. Photos can find duplicates inside your photo library, and Finder Smart Folders can filter by name and size, but macOS has nothing that scans a volume for exact copies and groups them.

Is it safe to delete duplicate photos?

Yes, when the match is byte for byte and one copy is kept. Be careful with edited versions and different resolutions of the same shot: those are not duplicates, they are different files, and a byte-for-byte comparison correctly leaves them alone.

How much space do duplicates usually take?

It varies wildly, but a few gigabytes is common and tens of gigabytes is not unusual on a Mac used for photography, video or software development. You only find out by measuring.