EzyTools

Character & Token Counter

Count characters, words, lines and bytes, with an approximate AI token estimate.

Loading…

Runs entirely in your browser. Nothing you enter is uploaded or stored on a server.

About this tool

Counting characters in Arabic is less simple than it looks: combining marks, emoji and diacritics make a naive count report more than a reader would. This tool counts characters the way a person sees them using Intl.Segmenter, shows the byte count that database and SMS limits actually use, and estimates tokens for anyone writing prompts for AI models.

How to use

  1. 1Paste or type your text into the field.
  2. 2Read the counts: words, characters, lines and paragraphs.
  3. 3Check the byte count if you have a technical limit.
  4. 4Use the token estimate when writing for an AI model.

When you need it

  • Staying within a character limit on a post, product description or ad.
  • Measuring the length of an article or an assignment in words.
  • Checking that text fits a database field limited in bytes.
  • Estimating the size of a prompt before sending it to an AI model.

Frequently asked questions

Why do characters and bytes differ?

An Arabic letter takes two bytes in UTF-8 and an emoji can take four, while a Latin letter takes one. Technical limits are usually in bytes, not characters.

Do diacritics count as characters?

The tool counts characters as a reader sees them, so a letter with a diacritic is one character, not two. The byte count does include them, because they are really there in the file.

How accurate is the token count?

It is an estimate. Models tokenise differently, and Arabic needs more tokens than English for the same meaning. Use it to size a prompt, not to predict a bill.

Related tools