Character Counter — SMS & Social Limits

SMS segments, explained properly

An SMS is not “160 characters” — it’s 140 bytes, and everything follows from how characters are packed into those bytes. The default packing, GSM-7, uses 7 bits per character: 140 bytes × 8 bits ÷ 7 = 160 characters. But GSM-7 only encodes a fixed alphabet from 1980s European telephony: unaccented Latin letters, digits, common punctuation, a handful of accented letters (é, ü, ñ, ö…), and some Greek capitals. Nine characters — € [ ] { } \ ^ | ~ and form feed — live in an extension table and cost two 7-bit slots each (an escape plus the character), which is why a message full of brackets runs out of room early; this page highlights them in amber as you type.

Any character outside all of that — an emoji, a smart quote “ ”, an em dash —, most accented letters, any non-Latin script — can’t be encoded in GSM-7 at all, so the entire message switches to UCS-2: 16 bits per character, 140 bytes ÷ 2 = 70 characters. One emoji doesn’t cost one character; it costs you the encoding. (Emoji outside the Basic Multilingual Plane, which is most of them, take two UCS-2 code units — so a single 🎉 uses 2 of your 70.)

When a message exceeds one SMS, it’s split into segments, and each segment gives up 6 bytes to a concatenation header telling the phone how to reassemble them. That shrinks per-segment capacity to 153 (GSM-7) or 67 (UCS-2). Carriers and SMS APIs bill per segment — a 500-character message with one emoji is 8 billable segments; the same message without it is 4. The segment counter above does this math exactly, including the extended-character double cost.

Platform limits

The table in the tool checks your text live against widely published limits: a post on X (280 characters for standard accounts), an Instagram caption (2,200) and bio (150), a LinkedIn post (3,000), a Discord message (2,000), a YouTube title (100), one SMS segment, an HTML title tag (~60 characters before search results truncate it), and a meta description (~160, though search engines actually truncate by pixel width, so ~160 is a guide rather than a wall). These numbers are the platforms’ generic public figures — they change at their owners’ discretion, and this site isn’t affiliated with any of them.

Counting rules: do emojis and URLs count differently?

Often, yes — a platform’s “character count” is frequently not text.length. Three cases worth knowing. Emoji: most emoji are multiple UTF-16 code units (and flag or family emoji are several codepoints joined invisibly), so naive counters over- or under-count them; some platforms count an emoji as 2, others as 1 grapheme. URLs: on X, every link is wrapped by the t.co shortener and counts as a fixed 23 characters regardless of its real length. Weighted characters: X counts most CJK characters as 2 toward its 280. This page reports honest Unicode counts (code points, with the SMS panel additionally doing code-unit math where that’s what’s billed) — treat platform-specific weighting as the platform’s own final word.

What each count means here

Characters counts Unicode code points — an emoji is 1, even when it’s 2 UTF-16 units (the SMS panel, where units are what’s billed, counts units). Without spaces excludes all whitespace. Words are runs of letters/digits. Sentences split on . ! ? followed by space or end — abbreviation-heavy text will over-count slightly, as every sentence counter does. Paragraphs are blocks separated by blank lines.

Frequently asked questions