Skip to content

SMS Character Limit: 160, 70 and Message Parts

A text message holds 160 characters, but only if every character is in the GSM 7-bit alphabet. One emoji, curly apostrophe or “á” switches the whole message to Unicode and cuts the limit to 70. Here is exactly why, and a counter that shows the encoding and number of parts.

Short answer

One SMS holds 160 characters in the standard GSM-7 alphabet, or 70 if the message contains any character outside it, such as an emoji, a curly quote or “á”. Longer texts are split into parts of 153 (GSM-7) or 67 (Unicode) characters, because each part carries a small header that tells the phone how to rejoin them.

Check your text message

0 characters

Assumes the standard GSM 7-bit alphabet. Phones and sending services sometimes swap characters before sending (Twilio’s Smart Encoding, for example, turns curly quotes into straight ones), which can change the result.

Counted on your device — nothing you type here is sent anywhere.

SMS limits at a glance

EncodingUsed whenOne SMSPer part when splitStandard maximum (255 parts)
GSM-7Every character is in the GSM alphabet16015339,015
UCS-2 (Unicode)Any character outside it706717,085

How many characters fit in each number of parts:

PartsGSM-7 charactersUnicode characters
116070
2306134
3459201
4612268
5765335
101,530670

A 161-character GSM message is two parts: 153 in the first and 8 in the second. Twilio notes one exception for its customers: multi-part messages sent from 8XX numbers (such as 800 or 888 numbers) to the US or Canada carry 152 GSM or 66 Unicode characters per part.

How SMS counts characters

140 bytes per message

The SMS standard (3GPP TS 23.040) gives each message 140 bytes of text. Everything else follows from how characters are packed into those bytes.

  • GSM-7: each character takes 7 bits. 140 bytes is 1,120 bits, and 1,120 ÷ 7 = 160 characters.
  • UCS-2: each character takes 16 bits (2 bytes). 140 ÷ 2 = 70 characters.

A message uses one encoding throughout. A single character outside the GSM alphabet moves all of it to UCS-2.

What is in the GSM alphabet

The GSM 7-bit default alphabet (3GPP TS 23.038) has 128 positions: A–Z, a–z, 0–9, space, line breaks, common punctuation, and a selection of accented and Greek letters chosen for European languages.

TypeCharactersCost
Standard GSMA–Z a–z 0–9 @ £ $ ¥ ¤ § ¿ ¡ _ ! " # % & ' ( ) * + , - . / : ; < = > ? é è à ù ì ò Ç É Ä Ö Ü ä ö ü Ñ ñ Ø ø Å å Æ æ ß Δ Φ Γ Λ Ω Π Ψ Σ Θ Ξ1 each
GSM extension table€ ^ { } [ ] ~ | \ and the page break2 each (escape code + character)
Forces Unicode’ ‘ “ ” (curly quotes) — – (long dashes) … • ` ° © ® ™, á í ó ú â ê ô ã õ, lowercase ç, tab, non-breaking space, any emoji, and letters from scripts such as Cyrillic, Arabic, Hindi or ChineseWhole message drops to 70 per SMS

We checked each character in the last two rows with the counter on this page, which follows the 3GPP tables. Note the oddities: “é” is in the alphabet but “á” is not, and capital “Ç” is in but lowercase “ç” is not.

The standard also defines national language tables (for Turkish, Spanish, Portuguese and several Indian languages such as Hindi, Bengali and Tamil) that let a phone send more letters in 7 bits. Support depends on the phones and networks involved, and they take extra header space, so this counter does not assume them.

Why parts hold 153 or 67

When a message is longer than one SMS, each part starts with a User Data Header of 6 bytes that numbers the parts so the phone can rejoin them. That leaves 134 bytes per part. TS 23.040 spells out the result: 153 GSM characters (160 − 7) or 67 Unicode characters ((140 − 6) ÷ 2). A two-position GSM character such as € is never split across two parts, and neither is a Unicode character.

Emoji take two units

UCS-2 was designed for 16-bit characters, and most emoji sit beyond that range. They are sent as two 16-bit units, so each uses 2 of the 70. Emoji with a skin tone, and flags, use 4. Twilio’s documentation confirms that an emoji consumes 2 characters of the limit.

Tips for shorter, cheaper texts

  • Hunt down curly quotes. Phones and word processors often turn ' into ’ automatically. One of them halves your limit. Swap them with Find and Replace before pasting into an SMS tool.
  • Check for invisible characters. A non-breaking space or tab copied from a document also forces Unicode. The Invisible Character Detector finds them.
  • Mind the cost per part. Services such as Twilio bill per message part, so a 140-character text with one curly quote costs two parts instead of one.
  • Use “...” and "-" in bulk messages. Three periods (3) and a hyphen (1) stay in GSM-7, while “…” and “—” switch the message to Unicode.
  • Watch the € and brackets. They stay in GSM-7 but count 2 each, so a price list full of € signs fills a message faster than it looks.
  • Remember the 160 is not universal. iMessage, RCS and chat apps such as WhatsApp don’t use SMS encoding, so these limits apply only when a message actually goes out as SMS.

For a plain count of characters and words, use the Character Counter.

Sources

Checked against these pages on . Platforms change their limits; if you spot a difference, tell us and we will re-check.

  1. 3GPP TS 23.038 (ETSI TS 123 038 V17.0.0) — Alphabets and language-specific information — GSM 7-bit default alphabet, extension table, national language tables
  2. 3GPP TS 23.040 (ETSI TS 123 040 V17.2.0) — Technical realization of SMS — 140 octets; concatenation: 153 (160 − 7), 67 ((140 − 6) ÷ 2), 255 parts
  3. Twilio Docs — How long can a message be? — 160 / 70, 153 / 67, the 152 / 66 exception, 1,600 maximum, billing per segment, Smart Encoding
  4. Twilio Docs — What is GSM-7 character encoding? — escaped characters count as two
  5. Twilio Support — Can I send and receive emoji with Twilio? — an emoji uses 2 characters of the 70

Frequently asked questions

How many characters fit in one text message?

One SMS holds 160 characters when every character is in the GSM 7-bit alphabet, which covers English letters, digits and common punctuation. If the message contains any other character, such as an emoji or a curly apostrophe, the whole message switches to Unicode and one SMS holds 70.

Why was my 100-character text split into two messages?

It almost certainly contained a character outside the GSM alphabet, often a curly apostrophe (’) added by autocorrect, an em dash or an emoji. That switches the message to Unicode, where one SMS holds 70 characters, so 100 characters need two parts.

Why are long texts split into 153 characters instead of 160?

Each part of a long message starts with a 6-byte header that tells the phone how to put the parts back together. Those 6 bytes take the room of 7 GSM characters, leaving 153. In Unicode they take the room of 3 characters, leaving 67 of 70.

Do emoji count as two characters in a text?

Usually, yes. An emoji first switches the whole message to Unicode (70 per SMS), and most emoji then take two 16-bit units each, so one emoji uses 2 of the 70. Emoji with skin tones or flags take 4.

Why does the euro sign count as two characters?

The € sign, the characters ^ { } [ ] ~ | \ and a page break live in an extension table of the GSM alphabet. Each needs an escape code plus the character, so it takes 2 of the 160.

What is the longest text message I can send?

The SMS standard allows up to 255 parts: 39,015 GSM characters or 17,085 Unicode characters. Phones and sending services set much lower limits; Twilio, for example, accepts up to 1,600 characters.

Tools for this

Character limits on other platforms