Free · Private · In your browser

Word Counter.

Live word, character, and line counts as you type — with dictionary-based segmentation for Japanese and Chinese, emoji counted the way you see them, an X post length check, and UTF-8 byte counts. Your text stays on your device.

Characters
0
No spaces
0
Words
0
Lines
0
Sentences
0
Paragraphs
0
Bytes (UTF-8)
0
Reading time

Manuscript pages

0 pages

Japanese 400-character genkōyōshi sheets, whitespace excluded

X post length

0 / 280

Weighted count: 1 for Latin, 2 for CJK and emoji. URL shortening (flat 23) not applied

Kanji
0
Hiragana
0
Katakana
0
Alphanumeric
0

Accepts .txt, .md, .yaml, .yml, and .json — files are read locally and never leave your browser. Your draft is saved automatically in this browser and restored the next time you open the page.

Features

Real-time statistics

Words, characters, lines, sentences, paragraphs, and UTF-8 bytes update instantly as you type or paste.

Built for Japanese and Chinese text

Dictionary-based segmentation (Intl.Segmenter) counts words in languages written without spaces, and every character count is script-aware — no more whole sentences counted as one word.

X post length check

See the weighted length X actually uses: Latin characters count as 1, CJK characters and emoji as 2, against the 280 limit.

Japanese manuscript-page conversion

Converts your character count into 400-character genkōyōshi sheets, the standard length unit for Japanese essays, submissions, and fiction.

Emoji-aware characters

Counts user-perceived characters (grapheme clusters), so a single emoji — even a family sequence — is one character, not two or eight.

Private by design

Everything runs in your browser. Your text is never uploaded, and drafts are auto-saved only on your own device.

How to use

  1. Type or paste your text into the sheet above.
  2. Or drag and drop a .txt, .md, .yaml, or .json file onto it.
  3. Read the counts — characters, words, manuscript pages, X length — as they update live.
  4. Use Copy to grab your text back; your draft is saved automatically.

Counting rules

Different tools disagree on counts because they define them differently. Here is exactly how this one counts.

Characters
User-perceived characters (grapheme clusters). An emoji, an accented letter, or a joined family emoji (👨‍👩‍👧) each count as one. Whitespace and newlines are included; a separate count excludes them.
Words
Counted with the browser's built-in Intl.Segmenter, which segments by dictionary. Japanese and Chinese text — written without spaces — is split into real words instead of being treated as one giant token. Punctuation-only runs are not words.
Lines
Editor semantics: the number of lines a text editor would display, split on LF, CRLF, or CR. A trailing newline opens a new line, so "a⏎" is 2 lines.
Sentences and paragraphs
Sentences use locale-aware segmentation on terminators like ".", "。", and "?". Paragraphs are blocks separated by blank lines; a single newline continues the same paragraph.
Bytes (UTF-8)
The byte length of the text encoded as UTF-8: ASCII is 1 byte per character, most CJK characters are 3, emoji are 4 or more. Useful when a form or database limit is defined in bytes.
Manuscript pages (原稿用紙)
Whitespace-excluded characters divided by 400, rounded up — the standard conversion used in Japanese submission guidelines. It does not simulate the extra sheets real page layout (paragraph breaks, dialogue lines) would add.
X post length
An approximation of X's weighted counting: most Latin-script characters and basic punctuation weigh 1; CJK characters, Hangul, and emoji weigh 2, against a limit of 280. The flat 23-character URL rule is not applied.

Questions

Why do other tools report a different count for the same text?

Tools disagree on how to treat spaces, newlines, and emoji. This tool counts user-perceived characters and shows counts both with and without whitespace, so you can match whichever definition your submission target uses. See the counting rules above for every definition.

How are Japanese and Chinese words counted?

With the browser's built-in Intl.Segmenter, which segments by dictionary rather than by spaces. That is why a Japanese sentence counts as several words here instead of one — most word counters get this wrong.

How accurate is the X post length?

The half-width/full-width weighting and emoji handling match X's official rules: Latin characters count 1, CJK and emoji count 2, limit 280. URLs are the exception — X counts any URL as a flat 23 characters, which this tool does not simulate, so posts with links show slightly high.

What is a manuscript page (genkōyōshi)?

Genkōyōshi is Japanese squared manuscript paper with 400 character cells per sheet. Assignment and contest length limits in Japan are often given in sheets ("within 5 pages"), so this tool converts your character count at 400 characters per page, whitespace excluded.

Is my text sent to a server?

No. All counting happens locally in your browser. Nothing you type or upload is transmitted or stored anywhere; the auto-saved draft lives only in your own browser's storage.

How do I delete the auto-saved draft?

Press Clear — it empties the sheet and removes the saved draft. Clearing your browser's site data also removes it.

Which file types can I upload?

You can drag and drop or select .txt, .md, .yaml, .yml, and .json files up to 5 MB. Files are read locally in your browser and never uploaded.