Skip to content
Sealpaste

Delete all recent work?

This removes the 0 items saved in this browser. Your settings stay.

Stop saving recent work?

New work won't be saved in this browser. The 0 items already saved can stay or go.

Character & Word Counter

Text

Paste or type text to count it

Counts
Ready
Runs in your browser. Nothing you paste is uploaded unless you share it.

Count characters, words, sentences and bytes as you type, with reading time, readability, keyword density and post limits

Paste or type text, or open a text, Word or OpenDocument file, to see its characters, words, sentences and lines, the counts code uses (code points, JavaScript’s length, UTF-8 and UTF-16 bytes), reading time, readability and the words used most. Set a limit, or pick one such as an X post, to see how many are left.

How to use it

  1. Paste or type the text, or open a text, .docx or .odt file.
  2. Read every count at once; each has its own Copy.
  3. Optional: pick a limit (an X post, an Instagram caption, a YouTube title…) or enter your own in characters, UTF-8 bytes or words to see how many are left or how far over it is.
  4. Read the words used most under Keywords, with stop words such as “the” and “of” left out in the language you pick.

Example

One emoji, four counts

Input
👨‍👩‍👧‍👦
Result
Characters: 1
Code points: 7
UTF-16 units (JavaScript length): 11
UTF-8 bytes: 25

A family emoji is four people joined by three zero-width joiners.

Which count does a limit use?

A limit is set in one unit, and an emoji or an accented letter is a different number in each. Check the one the system you write to counts in.

WhereCounts in
JavaScript, Java and C# string lengthUTF-16 units: an emoji outside the basic plane is 2
SQL VARCHAR(n)Code points in MySQL and PostgreSQL: n characters, whatever their bytes
Cookies, HTTP headers, many API fieldsBytes, usually UTF-8: é is 2, an emoji 4
A DNS label (one part of a domain name)63 bytes

Sources: MDN: String length, PostgreSQL: character types, RFC 1035 §2.3.4

What counts as one character

Characters here are what a reader sees as one: a flag, a skin-toned or joined emoji, or a letter with its accent marks is one, as Unicode’s grapheme rules define it. Code points count each part, so 👍🏽 is 1 character, 2 code points, 4 UTF-16 units and 8 UTF-8 bytes.

Words are found with the same rules browsers use to select a word: punctuation and spaces aren’t words, it’s and user_id are one word each, and Chinese, Japanese and Thai, written without spaces, are split by the browser’s dictionary.

So a hyphen and a slash split words: well-known is 2 words, user@example.com is 2 (user, example.com) and https://example.com/a/b?x=1 is 6. A counter that splits only at spaces says 1 for each; Microsoft Word says 1 for each too.

Sources: Unicode UAX #29

Readability and keywords

Flesch Reading Ease is 206.835 − 1.015 × (words ÷ sentences) − 84.6 × (syllables ÷ words): 60 to 70 is plain English, under 30 is hard. The Flesch-Kincaid grade turns the same two ratios into a US school grade: 0.39 × (words ÷ sentences) + 11.8 × (syllables ÷ words) − 15.59. Both are made for English; syllables are counted from vowel groups, as most counters do, so a word or two may be off by one.

Keywords are the words used most, with their share of all words (density). Stop words, the short words every text has, are left out in the language you pick: English, Spanish, French, German, Portuguese, Italian or Dutch, or none. Phrases are two or three words used together at least twice, not broken by a stop word or a comma.

Pages are the word count at 500 words a single-spaced page and 250 double-spaced, the usual rule of thumb for 12-point type.

Sources: Kincaid et al. (1975), the Flesch-Kincaid grade

Limits you can pick

Each preset sets the number and what it counts. X counts a link as 23 characters and Chinese, Japanese and Korean characters and emoji as 2, so a post with them reaches 280 sooner than this count says.

PresetLimit
X post280 characters
Bluesky post300 characters
Threads post, Mastodon post500 characters (a Mastodon server can set its own)
Instagram caption2,200 characters
LinkedIn post3,000 characters
YouTube title, description100 and 5,000 characters
Reddit title300 characters
Git commit subject50 characters, a convention, not a limit Git sets
DNS label63 bytes

Sources: X: counting characters, YouTube Help: add video details, Pro Git: commit guidelines

Sentences, paragraphs and reading time

Sentences are split by the same Unicode rules, so a full stop, question mark or exclamation mark followed by a space and a capital ends one; “e.g.” in the middle of a line may still count as an end. Paragraphs are blocks of text with a blank line between them; lines that only wrap stay in one paragraph.

Reading time is the word count at 238 words a minute by default, the average for silent reading of English non-fiction in a review of 190 studies; set your own speed beside it. Speaking time uses 183 words a minute, the same review’s pace for reading aloud. Both are averages: technical text reads slower.

Sources: Brysbaert (2019), How many words do we read per minute?

Questions

Is the text uploaded?
No. Everything runs in your browser, so your text stays on your device. Nothing you paste is uploaded unless you share it.
Can I count the words in a Word document?
Yes: open a .docx or .odt file and its body text is counted, paragraphs a blank line apart. Headers, footers, footnotes and deleted tracked changes are left out, as Word’s own count does by default. For a PDF, take its text out with TuckPDF’s PDF to Text (tuckpdf.com/pdf-to-text), our PDF site, and paste it here.
Are spaces counted as characters?
Yes: Characters counts spaces, tabs and line breaks. Characters without spaces leaves all of them out.
Why does JavaScript say my text is longer?
JavaScript’s length counts UTF-16 units. Emoji and some rare characters take two units each, and a joined emoji such as a family takes many, while a person sees one character.
How are lines counted?
Each line break (Windows CRLF, LF or CR) starts a new line, so text that ends with a line break has an empty last line. “Lines with text” leaves out empty lines and lines of spaces only.
Why is the UTF-8 count larger than the character count?
UTF-8 writes English letters, digits and ASCII punctuation in 1 byte, accented Latin, Greek, Cyrillic and Arabic letters in 2, most other scripts, Bangla and Chinese included, in 3, and emoji in 4.
What does “lone surrogate” mean?
Half of a UTF-16 pair with no other half, usually from text cut in the middle of an emoji. It can’t be written in UTF-8, so encoders replace it with �, which is 3 bytes.

Related tools