Exponents and math in plain text
Write x², x³, E=mc², or 2¹⁰ in any text field without an equation editor. The superscript digits ⁰–⁹ paste cleanly and render on every modern device, which makes them the most reliable part of the superscript set.
Text tool
Type any text below and get it in Unicode superscript (U+00B9, U+2070+), the block that gives you x²-style characters drawn smaller and higher than the baseline — for exponents, ordinals, and footnotes, without needing a font file installed.
Superscript text generated here is not a font. It is a different set of Unicode characters drawn smaller and higher than the baseline. The digits split across two ranges: ¹, ², ³ live in Latin-1 Supplement (U+00B9, U+00B2, U+00B3) for historical reasons, while ⁰ and ⁴–⁹ live in the Superscripts and Subscripts block (U+2070, U+2074–U+2079). Latin letters use modifier letters scattered across the Phonetic Extensions and Spacing Modifier Letters blocks — for example ᵃ is U+1D43, ⁿ is U+207F, ˡ is U+02E1. When you convert x2 into x², the app sees code point U+00B2 instead of the regular digit 2 — not a formatting instruction. The system font draws U+00B2 as a small 2 raised above the baseline. That is why superscript text survives copy-paste across platforms: the raised positioning lives in the character itself, not in a font you have to install.
Write x², x³, E=mc², or 2¹⁰ in any text field without an equation editor. The superscript digits ⁰–⁹ paste cleanly and render on every modern device, which makes them the most reliable part of the superscript set.
1ˢᵗ, 2ⁿᵈ, 3ʳᵈ, and reference markers like word¹ use superscript modifier letters. Useful in academic-style posts, Discord pinned messages, and anywhere you cite a source inline.
Discord chat, GitHub README, Notion, and plain email don't expose a superscript button. Pasting Unicode superscript characters works because they are just text — the raised positioning is baked into each code point.
Every superscript code point sits in the Basic Multilingual Plane, so each one costs 1 UTF-16 unit — not 2 like Mathematical Bold. A superscript-heavy caption barely changes length compared to plain text.
Every superscript code point lives in the BMP, so each costs 1 UTF-16 unit. Coverage is wide but uneven: the digits ⁰¹²³⁴⁵⁶⁷⁸⁹ are fully covered, and 24 of 26 Latin letters have a modifier-letter form that reads as superscript. The two gaps are q (no superscript form exists in Unicode) and a handful of letters that use Spacing Modifier Letters (U+02B0, U+02E1) rather than a dedicated superscript block, so their size varies by font. Operators work cleanly: superscript plus is U+207A, minus is U+207B, equals is U+207C, and parentheses are U+207D and U+207E. The legacy digits ¹²³ (U+00B9, U+00B2, U+00B3) live in Latin-1 Supplement and render slightly differently from ⁰ and ⁴–⁹ in some fonts — the height and weight do not always match. If exact visual consistency matters, avoid mixing the two digit ranges in one expression.
Superscript Unicode renders on iOS 13+, Android 8+, modern Windows, macOS, and every major browser. The digits ¹²³ (Latin-1) have near-universal support — even old devices render them. The U+2070+ digits and the modifier-letter superscripts (U+1D43, U+02B0, U+02E1) are covered by modern system fonts but render as boxes on a few older Android builds. Discord renders superscript reliably in chat, channel names, and custom status. GitHub shows it in README, issue, and PR bodies. Notion renders it in body text. TikTok accepts it in display names (30 chars) and bios (80 UTF-16 units), where the 1-unit cost helps. Instagram accepts it in the name field (30 chars) and bio (150 chars). Twitter/X accepts it in display names (50 chars) and tweets. Steam persona names (3–32 chars) accept it.
It is actually superscript. Each character is a separate Unicode code point (for example, ² is U+00B2, ⁿ is U+207F) that the system font draws smaller and higher than the baseline. You are not installing a font or applying formatting — you are pasting different characters that happen to be drawn as superscripts.
Unicode never assigned a superscript form for the letter q. Every other Latin letter has some modifier-letter equivalent that reads as superscript, but q is the one gap. The tool leaves q unchanged in the output because no usable substitute exists.
They come from different Unicode blocks. ¹²³ (U+00B9, U+00B2, U+00B3) were added early in the Latin-1 Supplement block for compatibility with legacy encodings. ⁰ (U+2070) and ⁴–⁹ (U+2074–U+2079) were added later in the Superscripts and Subscripts block. Some fonts draw the two groups at slightly different heights or weights.
Yes. The superscript digits ⁰–⁹ are well supported across modern devices, so x², x³, 2¹⁰, and E=mc² convert cleanly. This is the most reliable use case for superscript Unicode, especially the legacy digits ¹²³ which render even on older devices.
Yes, if you put it in page titles, meta tags, or any text you want indexed. Search engines normalize Unicode back to base letters but may not weight styled text the same way. Use superscript Unicode for display (captions, chat, READMEs), not for SEO-critical content.
Very little. Every superscript code point sits in the Basic Multilingual Plane, so each costs 1 UTF-16 unit — the same as a plain ASCII character. A 40-character superscript bio uses 40 of TikTok's 80 UTF-16 units.