Home / AI / Near Duplicate Finder

Near Duplicate Finder

Find lines, sentences or paragraphs that say the same thing, even in different words, and remove the copies.

Free, no uploads, no sign-up. Runs 100% in your browser.

How Near Duplicate Finder works

In AI mode each item becomes a MiniLM sentence embedding in a Web Worker and items closer than your threshold are grouped. Word overlap mode compares two-word shingles instead, with no download. The first item of each group is kept.

Input: Text, TXT, CSV. Output: TXT, CSV.

How to use Near Duplicate Finder

  1. Paste or open your text.
  2. Choose line, sentence or paragraph, and the similarity level.
  3. Find the groups.
  4. Download the cleaned text or the groups as CSV.

Tips

  • Nothing is uploaded. Your files stay in this browser tab.

Near Duplicate Finder FAQ

Are my files uploaded?
No. This tool runs entirely in your browser tab. Files are read from your disk into memory and the result is generated locally.
Is it free? Are there limits or watermarks?
Free and open source (MIT). There are no file-size caps, daily quotas, accounts or watermarks.
Does it work offline?
Yes, once the page (and any engine it needs) has loaded, it keeps working without a connection thanks to the service worker cache.

Related tools

  • Duplicate Line Remover: Remove repeated lines, keep only repeats, ignore case and spaces.
  • Semantic File Search: Search many PDFs, Word files, notes and code by meaning, not just exact words, all offline.
  • Zero-shot Text Classifier: Sort messages, tickets or headlines into labels you choose, with no training, on your device.
  • AI Summarizer: Summarize PDFs, Word files and long text on your device, as paragraphs or key points.

OpenFile runs in your browser. Turn on JavaScript to use the tools.