📊

Word Frequency Counter

Count how often every word appears, spot repeated phrases, and flag the words AI writing tends to overuse.

0Words
0Unique
0Sentences
0Characters
0mRead time
0Tokens ≈
#WordCountShare
Paste some text to see the counts.

Everything runs in your browser. Your text is never uploaded, which matters if you are checking a draft you have not published.

What people use it for

  • Checking a draft for words you have leaned on too heavily
  • Editing AI-assisted writing so it stops sounding like AI-assisted writing
  • Keyword density on a page you are optimising
  • Finding the phrase you unknowingly repeat in every paragraph
  • Estimating token cost before sending a long prompt to an API
  • Teaching — showing students their own repetition

Questions

Words and phrases that large language models reach for far more often than people do — delve, tapestry, seamless, "it is important to note", "in today's digital age". People use these too. What gives a text away is how many of them appear together.

No, and be wary of anything that says it does. Detectors are unreliable and regularly flag human writing. This counts specific overused words so you can reread those lines and decide for yourself.

No. Everything happens in your browser — nothing is sent to us, stored, or logged. You can disconnect from the internet and it still works.

Any run of letters, including hyphens and apostrophes, so "state-of-the-art" and "don't" each count once. Numbers and punctuation are skipped.

Otherwise the top of every list is "the", "and", "of". Switch it off if you want the raw counts.

It is a rough guide of about four characters per token, which is close for English prose. Every model tokenises slightly differently, so treat it as an estimate for budgeting, not a billing figure.

Looking for AI tools that actually work?

Every tool in our directory is opened and checked by a person before it goes live.

Browse the directory

More free tools

Free, no sign-up, nothing stored. Last updated August 2026.