Word Counter
Count words, characters, lines, paragraphs and bytes as you type. Free, no sign-up, runs in your browser.
FAQ
How is Chinese counted?
Chinese counts per character and Latin text per word, the usual convention for CJK writing. Character count with and without spaces plus byte size are shown too.
Which count do platform limits use?
Social platforms usually count characters, while academic submissions often specify characters too. When a limit is hard, follow the platform own counter.
What is the difference between bytes and characters?
One Chinese character takes 3 bytes in UTF-8 but counts as 1 character. Database column limits and SMS billing refer to bytes.
How many characters is an emoji?
Counting here is by Unicode code point, so one emoji is 1 character and 4 bytes. Some platforms count UTF-16 units and will show 2. When a limit is strict, trust the platform own counter.
About this tool
Eleven numbers at once: characters with and without spaces, words, lines, paragraphs, English words, UTF-8 bytes, plus counts of digits, letters, whitespace and punctuation, and a top-10 character ranking. Everything updates as you type and the text never leaves the browser.
"Words" means CJK characters plus Latin words. That is the usual convention for Chinese writing and closely matches Word and WPS: one Chinese character counts as one, and a run of Latin letters or digits counts as one. So "3 apples" is 2 by this metric while its character count is 8.
Characters are counted by Unicode code point, so one emoji is 1. Some platforms count UTF-16 code units and will show 2 for the same emoji. WeChat, Weibo and Xiaohongshu do not necessarily agree with each other, so when a limit is hard, trust the platform's own counter rather than this number.
Details that are easy to misread: lines equal newline count plus one, so a trailing newline adds an empty line; paragraphs are counted by blank-line separation, so in-paragraph line breaks do not start a new one; the English word matcher takes runs of letters and digits, ignores pure symbols, treats don't and co-op as one word, and counts a pure digit run as a word too.
Bytes are the UTF-8 length: three per Chinese character, four per emoji. Database column limits and SMS billing use this number rather than the character count (the classic 70-character SMS is really about encoded bytes). The top-10 ranking sorts visible characters by frequency, excludes whitespace, and treats fullwidth and halfwidth forms as different characters.