Unicode Versions Table
| Version | Date | Scripts | Characters | Headline additions |
|---|---|---|---|---|
| 1.0.0 | October 1991 | 24 | 7,129 | The first book: Arabic, Cyrillic, Greek, Hebrew, Hangul, Hiragana/Katakana and the other founding scripts |
| 1.0.1 | June 1992 | 25 | 28,327 | The first 20,902 CJK unified ideographs arrive in one drop |
| 1.1 | June 1993 | 24 | 34,168 | 4,306 new Hangul syllables; Tibetan removed (returned in 2.0); 33 characters reclassified as controls |
| 2.0 | July 1996 | 25 | 38,885 | The surrogate mechanism opens the 16-plane future; Hangul rebuilt in place (11,172 syllables); Tibetan readmitted |
| 2.1 | May 1998 | 25 | 38,887 | One headline glyph: the euro sign € |
| 3.0 | September 1999 | 38 | 49,194 | Cherokee, Geʸez, Khmer, Mongolian, Ogham, runes, Syriac, Thaana, braille patterns and the Yi syllables |
| 3.1 | March 2001 | 41 | 94,140 | Deseret, Gothic and Old Italic - the first dead-alphabet revivals - plus 42,711 CJK ideographs |
| 3.2 | March 2002 | 45 | 95,156 | The Philippine four: Buhid, Hanunoo, Tagalog (Baybayin) and Tagbanwa |
| 4.0 | April 2003 | 52 | 96,382 | Linear B and Cypriot syllabary (writing decrypted twice), Shavian, Osmanya, Ugaritic, Tai Le, Limbu |
| 4.1 | March 2005 | 59 | 97,655 | Glagolitic, Kharoshthi, Old Persian cuneiform, Tifinagh, Buginese, New Tai Lue, Sylheti Nagari |
| 5.0 | July 2006 | 64 | 99,024 | Sumero-Akkadian cuneiform lands whole; Balinese, NʻKo, Phags-pa, Phoenician |
| 5.1 | April 2008 | 75 | 100,648 | Lycian, Lydian, Carian, Vai, Saurashtra, Sundanese, Ol Chiki, Rejang, Cham, Kayah Li, Lepcha - and capital ị |
| 5.2 | October 2009 | 90 | 107,296 | The Sasanian wave: Avestan, Imperial Aramaic, Inscriptional Pahlavi, Inscriptional Parthian, Old Turkic, Old South Arabian, Samaritan, Egyptian hieroglyphs, Tai Tham, Tai Viet, Javanese, Kaithi, Lisu, Meetei Mayek, Bamum, Vedic extensions |
| 6.0 | October 2010 | 93 | 109,384 | Brahmi, Batak, Mandaic - and the first emoticons/emoji plus playing cards, traffic signs and alchemy |
| 6.1 | January 2012 | 100 | 110,116 | Chakma, Miao, Sharada, Takri, Sora Sompeng and both Meroitic scripts |
| 6.2 | September 2012 | 100 | 110,117 | One character: the Turkish lira sign ₺ |
| 6.3 | September 2013 | 100 | 110,122 | Five bidirectional formatting characters - plumbing, not glyphs |
| 7.0 | June 2014 | 123 | 112,956 | The largest script drop ever: Caucasian Albanian, Manichaean, Nabataean, Palmyrene, Psalter Pahlavi, Elbasan, Linear A, Grantha, Modi, Siddham, Tirhuta, Duployan, Mende Kikakui, Warang Citi, Pau Cin Hau and friends |
| 8.0 | June 2015 | 129 | 120,672 | Anatolian hieroglyphs, Ahom, Hatran, Multani, Old Hungarian, SignWriting, Cherokee lowercase and emoji skin tones |
| 9.0 | June 2016 | 135 | 128,172 | Adlam, Osage, Newa, Marchen, Bhaiksuki - and the Tangut corpus of nearly 6,000 characters |
| 10.0 | June 2017 | 139 | 136,690 | Zanabazar Square, Soyombo, Masaram Gondi, Nushu, Hentaigana - and the bitcoin sign ₿ |
| 11.0 | June 2018 | 146 | 137,374 | Old Sogdian and Sogdian, Hanifi Rohingya, Makasar, Medefaidrin, Dogra, Maya numerals and 145 emoji |
| 12.0 | March 2019 | 150 | 137,928 | Elymaic, Nandinagari, Wancho, Nyiakeng Puachue Hmong, small kana extensions and 61 emoji |
| 12.1 | May 2019 | 150 | 137,929 | One character: ㋿ Reiwa, the new Japanese era, rushed out on day one |
| 13.0 | March 2020 | 154 | 143,859 | Chorasmian, Dhives Akuru, the Khitan small script, Yezidi and 55 emoji |
| 14.0 | September 2021 | 159 | 144,697 | Delayed six months by COVID: Vithkuqi, Old Uyghur, Toto, Cypro-Minoan, Tangsa and 37 emoji |
| 15.0 | September 2022 | 161 | 149,186 | Kawi and Mundari Nag, 4,192 CJK ideographs and 20 emoji |
| 15.1 | September 2023 | 161 | 149,813 | A CJK repertoire extension plus the new CJK Extension H |
| 16.0 | September 2024 | 168 | 154,998 | Garay, Sunuwar, Gurung Khema, Kirat Rai, Ol Onal, Todhri, Tulu-Tigalari and 3,995 new Egyptian hieroglyphs |
| 17.0 | September 2025 | 172 | 159,801 | Beria Erfe, Tai Yo, Sidetic, Tolong Siki, the Saudi riyal sign and 4,316 CJK ideographs |
| 18.0 | September 2026 | 175 | 172,808 | Proto-cuneiform, Jurchen and Seal script - the biggest single jump (+13,007) since 3.1 |
Every app, font and operating system on earth inherits its characters from the release calendar below: thirty-one published versions of the Unicode Standard, from the 7,129-character first edition of October 1991 to version 18.0 (September 2026) with 172,808 characters and 175 scripts - a twenty-four-fold expansion that has run at a steady annual September rhythm since 2010.
The table is more than a changelog; it is a map of digital archaeology. The dead-alphabet revival project - the reason most historic scripts are typeable at all - began in 3.1 (2001) with Deseret, Gothic and Old Italic, and the 5.2 release of October 2009 admitted an entire Sasanian civilization in one drop: Avestan, Imperial Aramaic, Inscriptional Pahlavi, Inscriptional Parthian, Old Turkic, Old South Arabian and Samaritan. The emoji era started quietly in 6.0 (2010) and now anchors the whole schedule.
Reading the table: the Scripts and Characters columns are cumulative counts (the Wikipedia tabulation), so each row shows the state of the character universe after that release - and the Headline column names what each version is actually remembered for, from the euro sign’s solo debut in 2.1 to proto-cuneiform’s 13,007-character landing in 18.0.
How to use
- Find a character’s birth year: if you know which script or symbol you care about (say, Cherokee or the euro sign or emoji), scan the Headline column - each script appears exactly once, at the version that first admitted it, so the row gives you the minimum Unicode version any font or platform must support to render it.
- Use the version numbers as compatibility keys: software often advertises “Unicode 9.0+” or “Unicode 15.1 support” - read that as the table row (9.0 = 128,172 characters, June 2016, Adlam and Tangut) and you know exactly which characters are safe to ship.
- Watch the two counterintuitive columns: Characters is cumulative repertoire, not new additions (so 6.3’s five bidirectional characters is not a typo), and Scripts counts Wikipedia’s tabulation of writing systems - versions can add thousands of characters without adding a single script, and vice versa.
Frequently asked questions
What is the latest Unicode version?
Version 18.0, published September 2026: 175 scripts and 172,808 characters - the largest single-version gain (+13,007) since 3.1, led by proto-cuneiform, Jurchen and Seal script. New versions ship every September (with rare slips - 14.0 was delayed from March to September 2021 by COVID-19).
When was emoji added to Unicode?
Emoji proper arrived in version 6.0 (October 2010), which Wikipedia’s table headlines with “emoticons” alongside playing cards, traffic signs and alchemical symbols. Emoji then became the release calendar’s engine: nearly every version since has carried a headline emoji batch, and the skin-tone modifiers of 8.0 (2015) turned them into a combinatorial system.
Which version added [my script]?
Each script appears exactly once. Some landmarks: Cherokee in 3.0 (1999), Linear B in 4.0 (2003), Tifinagh in 4.1 (2005), cuneiform in 5.0 (2006), Avestan and Samaritan in 5.2 (2009), Brahmi in 6.0 (2010), Manichaean and Psalter Pahlavi in 7.0 (2014), Old Hungarian in 8.0 (2015), Adlam in 9.0 (2016), Sogdian in 11.0 (2018), Old Uyghur in 14.0 (2021) and proto-cuneiform in 18.0 (2026).
Why did version numbers jump from 13.0 to 14.0 a year apart?
They did not - but the dates moved. 13.0 shipped March 2020 on the old spring schedule; in April 2020 the Consortium announced 14.0 would slip six months to September 2021 because of the COVID-19 pandemic (Wikipedia cites this as the schedule’s only major break), and from then on September became the permanent release month.
How many characters does Unicode have now?
After version 18.0 (September 2026): 172,808 characters across 175 scripts, out of a codespace of 1,114,112 possible code points. The gaps are deliberate - roughly 880,000 code points remain unassigned, reserve space that has already shaped the standard’s surrogate and plane architecture since version 2.0 (1996).