How the duplicate file finder works
The tool reads each selected file in your browser and computes a SHA-256 hash. It groups files only when both their size and hash match, so a group contains exact, byte-identical files. File names do not determine a match.
Compare files
- Drop files into Drop files here to compare, or click browse files to choose multiple files.
- Wait while the tool hashes the files. The input has no configured file-type restriction.
- Use Add more files to extend the current comparison, or Clear all to remove the current selection.
Read the duplicate results
The summary cards show Files compared, Total size, Duplicate groups, Duplicate files, and Recoverable space. Duplicate files counts every file in a duplicate group. Recoverable space counts the copies beyond one file in each group.
The table shows Group, File name, Size, SHA-256, and Status. A group number identifies matching files; a dash marks a unique file.
Filter and export the results
Use Search files by name… to filter the visible rows. Select Duplicates only or Unique only to narrow the results. The count uses the format X of Y files. If both filters are active, no rows match.
Click CSV or JSON to export all processed files, including rows hidden by the current search or filters. CSV columns are name, size, sha256, group, and duplicate. JSON contains those same fields inside a files array. The tool does not delete, move, rename, compress, or otherwise modify files.
Processing limits and local behavior
No numeric maximum for file count, individual file size, or total data size is configured. Each file is loaded into an in-memory ArrayBuffer, so available browser or device memory can limit the practical batch size. Files are hashed sequentially. If reading or hashing fails, the current batch is not added and the tool shows an error beginning with Could not read a file:. Processing happens locally in the browser with the File API and Web Crypto API; nothing is sent to a server.
Frequently asked questions
Does this tool delete my duplicate files?
No. This tool only reports which uploaded files are byte-identical. It never deletes, moves, renames, or otherwise modifies any file - you decide what to do with the results yourself.
How does it decide two files are duplicates?
Each file is read and hashed with SHA-256. Files only land in the same duplicate group if both their size and hash match exactly - that is what "byte-identical" means here.
Is this the same as the Fuzzy Duplicate Finder?
No. This tool checks uploaded files for exact, byte-identical matches using size and content hashing. The Fuzzy Duplicate Finder is for near-match records in structured data rather than files.
Does it upload my files anywhere?
No. Files are read and hashed entirely in your browser using the File API and the Web Crypto API. Nothing is sent to a server.