DUPLICATE FILE LAB
Look beyond the name.
Find matching contents.
Narrow candidates by size, then compare SHA-256 hashes. Review content matches separately from matching names and export CSV.
01Choose files
Up to 1,000 files · 32 MiB each · 128 MiB total
Folder selection depends on browser support. Only files provided by the browser are scanned. Hidden files, symlinks and a complete disk inventory are not guaranteed. Only supplied relative paths are shown. Refreshing clears the input.
02Scan summary
256
Only files sharing a size with another file are read and hashed using Web Crypto SHA-256. Unique sizes are not hashed. Hash equality is strong evidence, but cryptographic collisions cannot be ruled out absolutely. Names and dates alone do not establish duplicate contents.
Exports both content-match and name-match groups. A file may appear in both categories.
03Results
No files selected. Load an example or choose files.
Scope and sources
All files stay on your device. No upload, original changes or deletion. Extra-copy size leaves one file per content group and sums the rest; actual recoverable disk space may differ.
Names are compared without case differences and with NFC normalization. Matching names do not imply matching contents.
Only files sharing a size with another file are read and hashed using Web Crypto SHA-256. Unique sizes are not hashed. Hash equality is strong evidence, but cryptographic collisions cannot be ruled out absolutely. Names and dates alone do not establish duplicate contents.
W3C Web Cryptography API