Text Statistics
Paste text and see how many words it has, whether it fits and by how much it overflows.
Text Statistics
Paste text and see how many words it has, whether it fits and by how much it overflows.
About Text Statistics
Chinese by character, English by word — counted separately
English word counters usually collapse a whole Chinese paragraph into a couple of words, and Chinese counters tend to swallow English as plain characters. Neither answers the question for text that mixes the two — a bilingual résumé, a deck with English product terms, a technical post. This tool reports the two numbers side by side so each one means what you expect.
Platform limits use each platform’s own counting rule
X does not count characters — it counts weighted units, and a Chinese, Japanese or Korean character costs two of them. So 280 units is roughly 140 Chinese characters, not 280. A tool that treats the limit as a plain character count will tell you there is room left when the post is already too long. The bar here applies the real rule, and marks which limits are approximate.
The same text always gives the same answer
Every rule here is written down explicitly: which code points count as Chinese characters, what makes a paragraph, how a line break is treated. Nothing depends on the browser’s built-in language dictionaries, which differ between browsers and versions and would quietly give two people different numbers for the same paragraph. Where a number is an estimate — reading time, SEO length — it says so on screen.
FAQ
Common questions and answers about this topic.
How are Chinese characters and English words counted?
Chinese characters are code points in the CJK ideograph blocks, counted one by one. English words are runs of letters and digits, where an apostrophe or hyphen inside a run keeps it as one word, so "don’t" and "state-of-the-art" each count as one. Characters are counted by code point rather than by UTF-16 unit, so a rare character such as 𠮷 counts as one character rather than two.
Why does the X count differ from other tools?
Because X counts weighted units, not characters. Latin letters and common symbols cost one unit each; everything else, including Chinese, Japanese and Korean characters, costs two. A 280-unit post therefore holds about 140 Chinese characters. Tools that treat the limit as 280 characters will say there is room left when the post would already be rejected.
How are lines and paragraphs decided?
A line is a run of text between line breaks, and a trailing line break does not add an extra line, so "a" followed by a newline is one line. A paragraph is a block separated from the next by at least one blank line, so a single line break does not start a new paragraph. Tools differ on both counts, which is why the rules are stated here.
How accurate is the reading time?
It is an estimate, not a measurement. The figure assumes a fixed pace — the rate is printed next to the result so you can judge it — and real reading speed varies with the reader, the subject and the layout. Treat it as a rough sense of scale for a talk or a post, not as a guarantee.
Why does Chinese word frequency show single characters?
Chinese is written without spaces, so finding word boundaries needs a segmentation dictionary. Without one, guessing the boundaries produces wrong words — splitting a technical term in the middle, for instance — and a wrong word list is worse than an honest character list. So Chinese is counted by character, consistent with how the main counts work.
Is my text sent anywhere?
No. Every number on this page is computed in the browser, and the text is not put into the address bar or saved between visits — close the tab and it is gone. That also means there is no share link for the text itself; use the copy button to take the numbers with you.