Character Count
Count characters, words, lines and bytes as you type. Handles Japanese and emoji correctly.
About this tool
Counting characters comes up constantly — checking the length of a draft, staying within a social post limit, or confirming a form's maximum input. This tool shows not just the character count but also words, lines, byte size, an estimate of 400-character manuscript pages, and approximate reading time, all updating live as you type. Your text is processed only in your browser and never sent anywhere.
The "character count" depends on how you count
It may be surprising, but there are several ways to count "characters," and the answer changes with the context. The three main ones are:
- Grapheme clusters — what a person perceives as one character. Emoji and accented letters count as one, matching the visible length you see on social media.
- Code points — the number of Unicode scalar values. Composite emoji are internally several code points, so they can count as more than they appear.
- UTF-8 bytes — the size as data. ASCII is 1 byte, most Japanese characters about 3, and some emoji 4 or more. Database column limits and API caps are often measured in bytes.
By default this tool counts by grapheme cluster, so emoji and combining characters match their visible length.
The emoji and combining-character trap
For example the family emoji 👨👩👧 looks like one character, but internally it joins several emoji with zero-width joiners, so a naive string length reports 3 or 7. The same applies to skin-tone-modified emoji and characters composed with combining marks (e.g. が = か + ゛). When a form's length check ignores these differences, you can get "too long" even though it looks like 10 characters. Being aware of the counting method explains such mismatches.
Matching a social or form limit
Different services count differently (grapheme, code point, or byte) and treat URLs and line breaks in their own way. To gauge a draft, the grapheme-based count (this tool's default) is closest to what you see. To match a strict limit exactly, check which unit the target service uses, and cross-check byte size as well to be safe.
Manuscript pages and reading time
To sense the volume of a draft, the tool also estimates pages on 400-character manuscript paper and an approximate reading time (a rough figure based on a typical reading speed). These are estimates, but they help when adjusting the length of a report or article, or judging how long a speech will run.
How to use
- Paste or type the text you want to count into the input field.
- Character, word, line and byte counts update as you type.
- Use “Copy result” to copy the summary.
FAQ
What is the difference between "characters" and "code points"?
Characters (graphemes) count what a person perceives as one character — emoji and combined characters count as one. Code points count Unicode’s internal units, which can exceed the character count for combined characters.
Is my text stored or sent?
No. All counting happens in your browser; your text is never sent to a server.
What are bytes useful for?
Program-side limits (database column sizes, payload sizes) are often measured in bytes, not characters. In UTF-8, one Japanese character is about 3 bytes.
How is the manuscript page count calculated?
It divides the character count by 400 and rounds up (a 400-characters-per-page convention). Spaces and line breaks count as cells too.
What is the reading time based on?
It approximates using roughly 500 characters per minute (a typical Japanese silent-reading speed). It is only an estimate and varies by content and reader.
How are words counted?
Runs of text separated by whitespace (spaces, line breaks, tabs) count as one word each, which suits English text.