Every Unicode Font Style Explained

Updated 8 August 2026 · about 1300 words

Copy-paste fonts are not fonts. They are characters borrowed from blocks encoded for mathematics, Japanese broadcasting and phonetic notation — which explains every gap and oddity you have run into.

There are no fonts here

Every "font generator" on the internet, this one included, is doing the same thing: swapping your letters for different Unicode characters that happen to be drawn in a different style. Nothing is downloaded, nothing is installed, and your text is still a single unformatted string.

That is precisely why it works in an Instagram bio, which offers no formatting whatsoever. And it is why the styles have gaps — you are limited to characters somebody had a reason to encode.

The Mathematical Alphanumeric Symbols block

Most styles come from one block, U+1D400–U+1D7FF, added in Unicode 3.1 in 2001. Its purpose was mathematical: a physicist writing 𝐅 = m𝐚 needs bold to mean vector and italic to mean scalar variable, and that distinction has to survive in plain text.

StyleExampleDigits?Notes
Bold serif𝐁𝐨𝐥𝐝YesComplete set
Italic serif𝐼𝑡𝑎𝑙𝑖𝑐NoLowercase h borrowed from elsewhere
Bold italic𝑩𝒐𝒍𝒅NoComplete letters
Script𝒮𝒸𝓇𝒾𝓅𝓉NoSeven capitals borrowed from an older block
Bold script𝓢𝓬𝓻𝓲𝓹𝓽NoComplete and consistent
Fraktur𝔉𝔯𝔞𝔨𝔱𝔲𝔯NoFive capitals borrowed
Bold Fraktur𝕱𝖗𝖆𝖐𝖙𝖚𝖗NoComplete
Double-struck𝔻𝕠𝕦𝕓𝕝𝕖YesSeven capitals borrowed
Sans-serif𝖲𝖺𝗇𝗌YesComplete
Sans bold𝗦𝗮𝗻𝘀YesComplete
Monospace𝙼𝚘𝚗𝚘YesComplete
Why some capitals are "borrowed": a handful of these letters had already been encoded years earlier as standalone mathematical symbols — ℋ for the Hamiltonian, ℝ for the real numbers. Unicode does not encode the same character twice, so those positions were left as holes and the earlier characters are reused. They were drawn by different designers, which is why they can look slightly out of step.

Enclosed Alphanumerics

Circled and squared letters come from U+2460–U+24FF and the Enclosed Alphanumeric Supplement at U+1F100. These were encoded for list markers, Japanese broadcasting symbols and content-rating badges.

  • Ⓒⓘⓡⓒⓛⓔⓓ — has both cases and digits; excellent support.
  • 🅒🅘🅡🅒🅛🅔🅓 filled — one case only, newer, sometimes rendered in colour.
  • 🅂🅀🅄🄰🅁🄴 squared — one case, no digits, and some pairs can trigger flag rendering because they share a block with regional indicators.

Halfwidth and Fullwidth Forms

Full-width letters (U+FF00–U+FFEF) exist because CJK typesetting places every character on a fixed square grid. A narrow Latin A breaks that grid, so a full-width duplicate was encoded.

Their association with 90s Japanese computing is what made them the signature of vaporwave typography. The block also contains the ideographic space, which is why the gaps between full-width words are so wide.

The incomplete sets

Three popular styles were never finished, because Unicode only encodes characters with a demonstrated need:

  • Small caps — added for phonetic notation, and it is missing a true small-capital Q and X. No digits either.
  • Superscript — nearly complete for lowercase, but there is no superscript q.
  • Subscript — only the letters chemists and mathematicians needed, so roughly half the alphabet is missing.

Every generator substitutes lookalikes for the gaps. That is a limitation of the standard, not of any particular tool.

Upside down and mirrored: pure improvisation

These two have no dedicated block at all. Unicode never encoded a rotated alphabet, so generators assemble one from wherever a suitable shape exists — the International Phonetic Alphabet, Greek, Cyrillic, mathematical operators, even CJK stroke characters.

That patchwork origin explains both quirks: some letters never look quite right, and coverage is uneven because the substitutes span so many blocks. Я, the reversed R used in mirror text, is simply a Cyrillic letter that happens to have the perfect shape.

See all the styles at once

Type one line and compare every style side by side.

Open the fancy text generator

More guides

Frequently asked questions

Are copy-paste fonts real fonts?

No. They are separate Unicode characters that look like styled letters. Nothing is installed, which is why they work in apps with no formatting options.

Why is there no small capital Q or X?

The small-caps letters were encoded for phonetic notation, and those two were never needed. Generators substitute lookalikes.

Why do some script capitals look different?

Seven of them were already encoded years earlier as standalone mathematical symbols. Unicode does not duplicate characters, so those earlier glyphs are reused.

Why are there no italic or script numbers?

Mathematics uses upright digits, so italic and script digit sets were never encoded. Bold, sans and monospace styles do include digits.

Where do upside-down letters come from?

There is no rotated alphabet in Unicode. Generators borrow shapes from phonetic, Greek, Cyrillic and CJK blocks that happen to look inverted.

Which style has the best device support?

Small caps, bold, italic, monospace, full-width and outlined circles are the safest. Filled circles and squares are newer and less consistently supported.