# X (Twitter) and LinkedIn character limits, explained

X (formerly Twitter) allows **280 weighted characters** in a standard post. LinkedIn's post limit is **3,000**, but it
counts text differently. Neither number tells you how much someone sees in their feed before expanding the post.
Try the [X character counter](/x-character-counter) or [LinkedIn character counter](/linkedin-character-counter) on your
final text; the differences below explain why a word processor's count may disagree.

Checked on October 26, 2026 against the platforms' own documentation, linked under [Sources](#sources). The examples
below were also checked against Flaghand's free counters and the scheduler's posting checks. Platform documentation
and measured preview behavior are called out separately.

## The limits at a glance

| Where you write | Limit | What that means |
|---|---|---|
| X standard post | 280 weighted characters | A recognized link uses 23; emoji and most Chinese, Japanese and Korean characters use 2. |
| X longer post with an eligible paid subscription | Up to 25,000 | Longer posts are a Premium feature, including Basic. Eligibility belongs to the posting account. |
| LinkedIn post | 3,000 | Spaces, line breaks and the full URL count. Most emoji use more than one unit. |
| LinkedIn comment | No verified general maximum in the checked docs | Ordinary-post Help and API docs give no number; group-specific Help pages disagree. |
| LinkedIn profile headline | Up to 220 | LinkedIn's own profile-writing guidance gives this number; it is not the post allowance. |

Flaghand's [X character counter](/x-character-counter) checks the standard 280 budget. The
[LinkedIn character counter](/linkedin-character-counter) checks posts, not comment or profile-headline limits.
Neither free tool verifies an account's paid subscription.

## X: why 280 does not mean 280 visible symbols

X documents the rules in [Counting characters](https://docs.x.com/fundamentals/counting-characters) and its
[twitter-text library](https://github.com/twitter/twitter-text):

- **Ordinary Latin letters, spaces and line breaks usually weigh 1.** Other scripts have their own Unicode ranges;
  this is not simply ASCII versus everything else.
- **Most Chinese, Japanese and Korean characters weigh 2.** A plain post of 140 such characters fills 280, before
  spaces, punctuation or links.
- **A recognized emoji sequence weighs 2**, even when several Unicode code points form one skin-toned emoji, flag or
  joined family.
- **Every recognized URL weighs 23**, whether it is shorter or longer than that. This is X's `t.co` treatment;
  manually shortening an already recognized link does not buy more room. Surrounding spaces still count.
- **X normalizes text to NFC before counting.** A decomposed `e` plus combining acute accent becomes the same `é`
  for this count. A word processor's character count can disagree.

Hashtags and manually typed `@handles` use their text's weight; they are not free. Automatically included reply
mentions are a separate X rule, not a reason to omit an `@handle` you typed from the count.

`twitter-text` explains the underlying model, but a particular library release is not a guarantee of today's X
composer behavior. Flaghand shares its X counting code between the free counter and the desk's checks, following
the current composer rules rather than treating JavaScript's string length as X's count.

## Premium and longer X posts

[About X Premium](https://help.x.com/en/using-x/x-premium) lists longer posts up to **25,000 characters**, with
longer posts included in Basic, Premium and Premium+. Its official search-indexed excerpts confirmed those details;
the complete Help page was blocked to our reader on the check date. That does not turn every account's post limit
into 25,000, or make a free counter's negative “left of 280” result proof that a paid account cannot post it.

For a single longer post, Flaghand checks the connected account's subscription type from X and applies a **25,000
weighted-character** ceiling when X confirms `Basic`, `Premium` or `PremiumPlus`. An editable blue badge in a preview
is not that confirmation. X's [API changelog](https://docs.x.com/changelog) announced long-form writes on August 9,
2024 and an increase from 4,000 to 25,000 on August 20, 2024. Those entries establish API support, not an entitlement
guarantee for every app and account.

Longer-post eligibility and how much the timeline shows are different things. Preview the opening rather than
assuming everyone sees all 25,000 characters without expanding it.

## LinkedIn: the post budget is not X's weighting

[LinkedIn's post-versus-article Help page](https://www.linkedin.com/help/linkedin/answer/a522483/differences-between-posting-updates-and-publishing?lang=en) documents a **3,000-character
post** allowance. Flaghand's counter and scheduler count the plain text as
posted in **UTF-16 code units**, matching the composer behavior covered by its counting fixtures:

- An ordinary Latin letter, a space, a line break or a basic Chinese, Japanese or Korean character counts 1.
- A common emoji such as `😀` counts 2. A skin tone, flag or joined family can use more.
- A URL counts its **full typed length** toward the post budget, even when the feed displays a shortened link.
- Mathematical Unicode bold or italic letters often count 2 each. They are different characters, not rich-text
  formatting; the [LinkedIn formatting guide](/docs/guides/formatting-a-linkedin-post) explains the accessibility and
  search trade-offs, as well as spacing, mentions and links.

The 3,000 allowance is LinkedIn's published rule. The UTF-16 explanation and the examples here describe the counting
behavior Flaghand implements and tests; they are not a claim that LinkedIn's help page specifies its Unicode algorithm.

## Comments and headlines have separate rules

LinkedIn's [ordinary-post commenting Help page](https://www.linkedin.com/help/linkedin/answer/a524166/commenting-on-updates-and-comments?lang=en) and current
[Comments API reference](https://learn.microsoft.com/en-us/linkedin/marketing/community-management/shares/comments-api?view=li-lms-2026-10)
do not specify a numeric comment-text maximum. Its group-specific Help pages disagree:
[General limits for LinkedIn groups](https://www.linkedin.com/help/linkedin/answer/a544801?lang=en) lists **1,250 for
comments**, while [Add comments to group conversations](https://www.linkedin.com/help/linkedin/answer/a1651958?lang=en)
says **1,000**. Neither is a verified general limit for ordinary feed comments or the API. Check the actual comment
field's feedback rather than borrowing the post's 3,000 allowance or treating a group number as universal.

For a **personal profile headline**, LinkedIn's own
[profile-writing guidance](https://business.linkedin.com/sell/resources/effective-sales-profile) says headlines can be
**up to 220 characters**. That is the headline under your name, not a post, article title or ad headline. Its
[headline Help page](https://www.linkedin.com/help/linkedin/answer/a542926/editing-your-professional-headline?lang=en)
explains how to edit it but does not state a number. Flaghand's free post counter does not validate either comments
or profile headlines.

## What the same text counts

Each row is just the text inside the code marks; the marks do not count.

| Text | X | LinkedIn |
|---|---|---|
| `A` | 1 | 1 |
| `漢` | 2 | 1 |
| `😀` | 2 | 2 |
| `👍🏽` | 2 | 4 |
| `👨‍👩‍👧‍👦` | 2 | 11 |

Links make the difference especially clear:

- `https://example.com`: **23 on X, 19 on LinkedIn**.
- `https://example.com/a/long/path`: **23 on X, 31 on LinkedIn**.

One more edge case: visually identical accents need not have the same LinkedIn count. The single character `é`
counts 1; an `e` followed by a combining acute accent counts 2. X normalizes both to a weight of 1.

## How much appears before “see more”?

**It is not a fixed character allowance.** The post can fit its platform's posting limit while part of it is hidden
in the feed. Width, line breaks, glyph widths and the particular feed view affect what appears.

The LinkedIn Help and API pages checked here do not promise one universal “first 140” or “first 210 characters” cutoff. In
Flaghand's measured LinkedIn feed renderer, a **552 px** desktop card starts with a three-line clamp, dropping
trailing empty lines from that preview. A blank third line can therefore leave only two visible lines. A narrower
phone card wraps the same words sooner. This is a measured preview model, not a LinkedIn API contract.

Use the [LinkedIn character counter](/linkedin-character-counter) to shade what falls behind the fold, then open the
[LinkedIn post previewer](/linkedin-post-preview) to check the actual desktop and phone layouts. The
[formatting guide](/docs/guides/formatting-a-linkedin-post) covers the 552 px behavior in more detail.

## Before publishing

Count the final text after adding links, emoji and styled letters. Then inspect its opening at the width readers
will use. In the hosted desk, an agent can draft, check and request approval; only the owner approves the exact
snapshot with a passkey. No agent token can approve, schedule or post. [Get access to Flaghand](/schedule).

## Sources

Platform pages checked on October 26, 2026. No real post, comment, profile edit or paid API call was made for this guide.

- X: [Counting characters](https://docs.x.com/fundamentals/counting-characters) and the platform-owned
  [twitter-text library](https://github.com/twitter/twitter-text).
- X longer posts: [About X Premium](https://help.x.com/en/using-x/x-premium) (official indexed excerpts; direct Help
  access blocked on the check date) and the [API changelog](https://docs.x.com/changelog) (August 9 and 20, 2024 entries,
  directly read).
- LinkedIn posts: [Differences between posting updates and publishing articles](https://www.linkedin.com/help/linkedin/answer/a522483/differences-between-posting-updates-and-publishing?lang=en).
- LinkedIn comments: [Comment on posts and reply to a comment](https://www.linkedin.com/help/linkedin/answer/a524166/commenting-on-updates-and-comments?lang=en)
  and the [Comments API](https://learn.microsoft.com/en-us/linkedin/marketing/community-management/shares/comments-api?view=li-lms-2026-10).
  Conflicting, group-specific numbers: [General limits for LinkedIn groups](https://www.linkedin.com/help/linkedin/answer/a544801?lang=en)
  and [Add comments to group conversations](https://www.linkedin.com/help/linkedin/answer/a1651958?lang=en).
- LinkedIn headline: [LinkedIn's profile-writing guidance](https://business.linkedin.com/sell/resources/effective-sales-profile)
  and [Edit your headline](https://www.linkedin.com/help/linkedin/answer/a542926/editing-your-professional-headline?lang=en).

The numerical examples, free-tool behavior and posting checks were compared with Flaghand's shared counters and
scheduler. The “see more” section uses Flaghand's measured renderer and is deliberately not presented as a fixed
limit from LinkedIn's documentation.