Haru ToolsHow to use this tool
Different writing limits measure different things. A word limit, a character limit, and a byte limit are not interchangeable, especially when text contains emoji or multiple languages. Paste or type text to view several measurements together. The character count uses grapheme clusters, the text segments that generally correspond to a character as a reader perceives it.
- Paste or type the text you want to measure.
- Choose the measurement that matches your requirement: characters, characters without whitespace, word groups, lines, or UTF-8 bytes.
- For a specific publishing platform, check its own final counter because it may define or weight characters differently.
The calculation, explained
Hello world
11 characters, 10 without whitespace, 2 word groups, 1 line, 11 UTF-8 bytes
The space counts as a character and a byte, but separates the two words. These ASCII letters each occupy one UTF-8 byte; other scripts and emoji often need more.
A visible character can contain several code points
An accented letter can be stored as one Unicode code point or as a letter followed by a combining mark. Some emoji join several code points into one visible symbol. Grapheme counting aims to keep those sequences together, so it can differ from a programming language's string length.
Words and lines use practical boundaries
Word groups are non-empty groups separated by whitespace. An unspaced Chinese, Japanese, or Korean sentence therefore counts as one group, not its linguistic word count. Lines are separated by line breaks; a final line break creates a final empty line. Empty text has zero lines.
Bytes describe an encoding, not visual length
The byte total is for UTF-8 encoding of the text itself. It does not include a document container, formatting, an attachment, or network headers. The no-whitespace count removes spaces, tabs, and line breaks; punctuation remains. A service may normalize text, count links specially, or apply its own limit.
Questions that come up
Is an emoji always one character?
A single displayed emoji is often one grapheme cluster, even when it contains several code points. Byte counts can still be much larger. Display and segmentation support may vary with the browser's Unicode version.
Does the word count work like a Korean or Japanese dictionary?
No. It counts groups separated by whitespace. Korean spacing units and unspaced Japanese or Chinese text are not the same as linguistic word counts.
Why does another website show a different count?
It may count code units, code points, bytes, weighted characters, or normalized text. Use its published definition and its own submission counter when meeting a hard limit.
Reviewed: September 21, 2026 · Haru Tools editorial