Character Counter

About the Character Counter

Type or paste text and see its length counted several ways at once: characters with and without spaces, words, lines, UTF-8 bytes and UTF-16 code units. Bars show how much room is left for an X post, an SMS, a title tag, a meta description, an Instagram caption or a YouTube title, and for a limit you set yourself.

A character here is what you see as one symbol, so 👍🏽 or a flag counts as 1 even though it is stored as several code points. When the text is too long, the part that does not fit is highlighted, so you know exactly where to cut.

Common uses

• Fitting posts, captions and bios into social media limits.
• Keeping SMS campaigns and OTP templates in one message and in the cheaper GSM-7 encoding.
• Checking title tags and meta descriptions before publishing a page.
• Making sure text fits a database column, an API field or an HTML maxlength measured in bytes or code units.
• Finding invisible characters (zero-width spaces, byte order marks) that sneak in with copy and paste.

How X (Twitter) counts characters

X does not simply count characters. Its open-source twitter-text library, whose rules this counter follows, gives each character a weight:

• Letters, digits and punctuation from Latin, Greek, Cyrillic, Hebrew, Arabic, Devanagari and the other scripts up to U+10FF, plus common spaces, dashes and quotation marks, count as 1.
• Chinese, Japanese and Korean characters and most symbols count as 2, so a post in those scripts holds about 140 characters.
• Every emoji counts as 2, including skin-tone, flag and family sequences.
• Every link counts as 23, however long or short it is, because X wraps links in its t.co shortener. That includes links typed without http://, such as toolverse.in or example.com/blog, as long as they end in a real domain ending; e-mail addresses are not links.

Text is normalised (Unicode NFC) first, so an accent typed as a separate mark is not counted twice. These rules are described in X’s developer documentation. The 280 limit applies to standard accounts; X Premium subscribers can post longer text.

SMS: GSM-7, Unicode and message parts

Text messages use one of two encodings, defined in the 3GPP specifications TS 23.038 (alphabets) and TS 23.040 (long messages):

• GSM-7 covers English letters, digits, common punctuation and some accented letters (é, è, ü, ñ, Ä, Ö…). One message holds 160 characters. The characters ^ { } [ ] ~ \ | € need an escape code, so each counts as 2.
• Unicode (UCS-2) is used as soon as the message contains any other character — an emoji, a curly apostrophe (’), an en dash (–), Hindi or Chinese text. One message then holds only 70 characters.

Longer texts are split into parts that the phone joins back together. Each part gives up some room to a header, so a part holds 153 GSM-7 or 67 Unicode characters. Most providers bill each part as a separate SMS.

Curly quotes and dashes pasted from a word processor are the most common reason a message unexpectedly becomes Unicode. Replace look-alikes swaps them for plain equivalents without changing any letters.

Platform limits used here

Each limit comes from the platform’s own documentation:

• X post: 280, weighted as described above (twitter-text configuration).
• Instagram caption: 2,200 characters, at most 30 hashtags and 20 @ mentions (Instagram API reference). The counter shows how many hashtags and mentions your text has.
• YouTube: title 100 characters, description 5,000 (YouTube Help).
• Threads post: 500 characters, with each emoji counted as its UTF-8 size, so 😀 uses 4, and at most 5 different links (Threads API documentation). Threads does not say how it spots links, so the counter uses X’s rules.
• LinkedIn post: 3,000 characters (LinkedIn Help).
• Bluesky post: 300 characters as you see them (graphemes) and at most 3,000 bytes in UTF-8 (AT Protocol post schema). The byte limit matters for long posts in scripts such as Hindi, where one character can take 6 bytes or more.

Title tags and meta descriptions

Google does not publish a character limit for either. It shortens title links and snippets to fit the width of the screen, so the 60 and 160 characters used here are guides, not rules: wide letters such as W and M take more room than i or l. Put the most important words first so nothing that matters is cut off.

Characters, code points, UTF-16 units and bytes

The same text has several lengths, and different systems use different ones:

• Characters: what you see. 👍🏽 is 1 character.
• Code points: Unicode values. 👍🏽 is 2 (thumbs up + skin tone). PostgreSQL and MySQL (utf8mb4) measure VARCHAR lengths this way.
• UTF-16 code units: how JavaScript, Java and the HTML maxlength attribute measure text. 👍🏽 is 4.
• UTF-8 bytes: the storage size in files, URLs and most APIs. An English letter is 1 byte, é is 2, a Hindi letter is 3 and 👍🏽 is 8.

Limitations

• Instagram, YouTube and LinkedIn limits are counted in characters as you see them. Their documentation does not say how emoji are counted, so leave a few characters spare if your text has emoji.
• Links are found with the rules and the list of domain endings in X’s open-source twitter-text library, whose list does not include the newest domain endings; a link to a domain ending missing from it may be counted differently in X.
• Every emoji counts as 2 here, as X’s documentation describes. The emoji list inside X’s open-source library is not kept up to date, so tools built on it count some newer emoji made of several code points (such as 🧑‍🧑‍🧒) as more; if a post with very new emoji is close to the limit, check it in X itself.
• SMS counts assume the standard GSM-7 alphabet. Some networks also support national language tables (for example for Turkish or Indian languages), which this counter does not model.
• Platforms change their limits from time to time. If a post is rejected, check the platform’s own help pages.

Privacy

Everything happens in your browser. What you enter or open here is not uploaded or stored by MySmartCoPilot.