Bluesky Character Limit 300: How Hangul and Emojis Count

The text of a Bluesky post must be at most 300 visible characters (graphemes) and at most 3,000 UTF-8 bytes. Both conditions must hold. This article uses Korean text as its multi-byte example, so Korean example sentences are kept as written. One Korean syllable is 1 character and 3 bytes, so 300 Korean characters are 900 bytes and only the character limit matters. An emoji counts as one character but takes many more bytes, so a post with many joined emojis such as the family emoji hits the byte limit first. The values below come from the public Bluesky lexicon, checked with the Geuljasu count functions.

Woman typing a short post on her smartphone by a cafe window AI-generated image

How many characters can one Bluesky post hold?

The “app.bsky.feed.post” lexicon in Bluesky’s public atproto repository on GitHub defines the post text (text) like this.

The two numbers use different units. According to the AT Protocol specification, maxGraphemes counts Unicode grapheme clusters, meaning what a person sees as one character, while maxLength counts UTF-8 bytes. A post must stay within both limits at once. The lexicon also notes that the text may be an empty string when there are embeds, so a post without written characters is possible.

The profile lexicon (app.bsky.actor.profile) sets the display name at 64 characters and 640 bytes, and the description at 256 characters and 2,560 bytes. The extra tags field on a post allows up to 8 tags, each 64 characters and 640 bytes.

FieldCharacters (graphemes)Bytes (UTF-8)
Post text3003,000
Profile display name64640
Profile description2562,560
One post tag64640

One thing stands out: the byte limit is exactly 10 times the character limit in every field. If the average character takes more than 10 bytes, the byte limit fills first. If it takes 10 bytes or fewer, the character limit fills first.

What is the difference between a character (grapheme) and a byte?

A grapheme is what a person sees as one character. A Korean syllable, a letter, a digit, and a single emoji are each one grapheme. A byte is the size a computer uses to store that character, and it varies a lot by type even when the character count is the same. Filling 300 characters with different kinds of characters and counting them with the Geuljasu functions gives this.

UTF-8 bytes taken by the same 300 characters300 English letters: 300bytes; 300 Hangul: 900bytes; 300 smiling-face emojis: 1,200bytes; 300 flag emojis: 2,400bytes; 300 family emojis: 7,500bytes02,0004,0006,0008,000300 Englishletters300bytes300 Hangul900bytes300 smiling-faceemojis1,200bytes300 flag emojis2,400bytes300 familyemojis7,500bytes
UTF-8 bytes taken by the same 300 characters — The post text byte limit is 3,000 (values calculated directly)

All of them are 300 characters, yet the bytes range from 300 to 7,500. English is 1 byte, Hangul is 3 bytes, and a basic emoji is 4 bytes. A flag emoji joins two characters for 8 bytes, and a family emoji combines four person emojis and three joiners for 25 bytes. It looks like one emoji, but it is 1 character and 25 bytes.

How many Korean characters can I post?

One Korean syllable is 1 grapheme and 3 bytes in UTF-8. Counting 300 repeats of the syllable 가 (ga) gives 300 characters and 900 bytes. That fits the character limit of 300 exactly and leaves plenty of room under the 3,000-byte limit. One more character makes 301, which is still within the byte limit at 903 bytes but over the character limit, so it cannot be posted.

By bytes alone, 1,000 Korean characters (3,000 bytes) would fit, but the character limit of 300 stops you first. If you write only Korean, ignore bytes and watch just the 300-character count, spaces and line breaks included. You can check the count with spaces in the Word Counter.

Which limit do I hit first when I add emojis?

An emoji is one character, so it barely affects the character limit. The problem is bytes. The table below uses values counted with the Geuljasu functions.

Post contentsCharactersBytesResult
300 Hangul300900Pass
301 Hangul301903Over the character limit
300 smiling-face emojis3001,200Pass
300 flag emojis3002,400Pass
120 family emojis1203,000Pass (equal to the limit)
121 family emojis1213,025Over the byte limit
204 Hangul + 96 family emojis3003,012Over the byte limit
205 Hangul + 95 family emojis3002,990Pass

Putting more than 120 family emojis in a post is rare, but a post like “204 Hangul + 96 family emojis” has exactly 300 characters and still goes 12 bytes over. You get blocked despite a fine character count when a post has many characters made of several joined emojis. A few basic emojis such as a smiling face or a heart barely change the bytes.

This table checks the numbers in the Bluesky lexicon against the Geuljasu functions. Whether the remaining-character display in the app input screen works the same way can depend on the app version, so check once more on the real screen. How text is split into characters (Intl.Segmenter) can also differ for newer emojis depending on the browser’s Unicode version.

In what order should I check before posting?

Checks before posting on BlueskyWrite post → 300 chars or fewer?; 300 chars or fewer? → 3,000 bytes or fewer? (Yes); 300 chars or fewer? → Trim text or emojis (No); 3,000 bytes or fewer? → Post (Yes); 3,000 bytes or fewer? → Trim text or emojis (No); Trim text or emojis → 300 chars or fewer?YesNoYesNoWrite post300 chars or fewer?3,000 bytes orfewer?PostTrim text or emojis
Checks before posting on Bluesky — Both the character count and the bytes must pass

For a post that is mostly Korean, the first question is enough. If you joined several emojis or mixed in other scripts, check the bytes at the second question too. When over, trim text or emojis, whichever limit you passed. To see where text goes past a limit, put the limit into the Character Limit Trimmer and paste your text.

Code to check it yourself

Paste this into the browser console to see the character count and UTF-8 bytes together. It uses the same rules as the values in the table above.

// Count graphemes and UTF-8 bytes together to check both limits
function blueskyCount(text) {
  const seg = new Intl.Segmenter('ko', { granularity: 'grapheme' });
  const graphemes = [...seg.segment(text)].length;
  const bytes = new TextEncoder().encode(text).length;
  return { graphemes, bytes, ok: graphemes <= 300 && bytes <= 3000 };
}
const family = '\u{1F468}\u200D\u{1F469}\u200D\u{1F467}\u200D\u{1F466}';
blueskyCount('가'.repeat(300));                       // 300 chars, 900 bytes, pass
blueskyCount('가'.repeat(204) + family.repeat(96));   // 300 chars, 3012 bytes, over

Frequently asked questions

Whatever you write in the post, whether an address or a mention, is part of the post text and is counted the same way. We could not confirm how links might be shortened or handled separately. Leave some room when a post contains a link.

Are line breaks and spaces counted?

They are characters in the post text, so by definition a line break counts as one character. With the Geuljasu functions, one line break is 1 character and 1 byte, and a Windows-style line break (CRLF) is 1 character and 2 bytes. The app display may differ, so leave some room near the limit.

If I write only Korean, can I ignore bytes?

300 Korean characters are 900 bytes, well under the 3,000-byte limit. Unless you add many joined emojis or a lot of other scripts, only the 300-character count matters. You can check bytes on the UTF-8 basis in the Byte Counter.

Basis: Bluesky atproto repository lexicons "app.bsky.feed.post" and "app.bsky.actor.profile" (text maxGraphemes 300, maxLength 3000), AT Protocol Lexicon specification (maxLength is UTF-8 bytes, maxGraphemes is graphemes), Geuljasu character and byte count functions. Sources checked on 2026-11-14. Byte values in this post are calculated the same way as the Geuljasu tools; real SMS services may follow different rules set by carriers and sending providers.

The full counting rules are on Methods and standards.

Tools related to this post