Small capitals read as emphasis without shouting, which is why they suit a bio line. These are real unicode characters from the phonetic block, so they paste anywhere.
Type into the small caps generator and each lowercase letter is replaced by a different Unicode character that happens to be drawn as a miniature capital. Nothing about the font changes. The letter a becomes ᴀ, U+1D00 LATIN LETTER SMALL CAPITAL A, a distinct codepoint with its own name, its own character properties, and no formal relationship to the a you typed. This is codepoint substitution, the same mechanism behind every style on this site, and it is why the result survives a copy and paste into a field that carries no formatting at all. The trade is that you are no longer sending the word you typed. You are sending twenty-six lookalike characters that a machine treats as unrelated symbols. The 33 styles here differ in what they mix in around that core mapping: some pair small capitals with spaced letters, some with brackets or separators, some fall back to full capitals for letters where no small capital exists. The substitution runs entirely in your browser; nothing you type is sent anywhere.
These glyphs were not designed for styling. Most come from the International Phonetic Alphabet, where a small capital is a distinct symbol for a distinct sound: ɢ (U+0262) is a voiced uvular plosive, ʀ (U+0280) a uvular trill, ɴ (U+0274) a uvular nasal, ʙ (U+0299) a bilabial trill. Linguists needed them in print and then on screen, so Unicode encoded them, the IPA Extensions block (U+0250–U+02AF) in version 1.0, then the Phonetic Extensions block (U+1D00–U+1D7F) in Unicode 4.0 (2003) for the remainder, and a handful more in Latin Extended-D (U+A720–U+A7FF) from Unicode 5.0 (2006). Typographic small caps are a much older idea, running back to Renaissance printers who cut a second set of capitals at roughly x-height with thickened strokes so they would not look spindly beside lowercase. Those live inside a font file as alternate glyphs, selected by the OpenType smcp feature. Unicode deliberately never encoded them, because they are a style of the same letter, not different letters. The IPA symbols are what people reached for instead, and that borrowing is what you are using here.
The alphabet is incomplete, and two letters give it away. There is no small capital x in Unicode at all, no proposal ever justified one, because no phonetic notation needed it. Generators substitute either a plain lowercase x, which sits visibly shorter and lighter than its neighbours, or ˣ (U+02E3), a modifier letter meant for superscripts, which floats above the baseline. Q was missing until U+A7AF (ꞯ) arrived in Unicode 11.0 in 2018, far too late for most installed fonts to have picked it up; on many devices it renders as tofu, the empty rectangle a renderer draws when nothing in the font stack can supply a glyph. F and S are also outliers: ꜰ (U+A730) and ꜱ (U+A731) live in Latin Extended-D rather than the phonetic blocks, so a font can support twenty-four letters and still miss those two. The practical result is that a word containing q or x will look wrong somewhere, and there is no fix in the encoding, only a choice of which compromise you prefer.
Coverage follows the linguists. Segoe UI on Windows, San Francisco on Apple platforms, Roboto and the Noto families on Android and DejaVu on Linux all carry the IPA Extensions and Phonetic Extensions blocks, because academic and dictionary typesetting demanded them long before anyone used them decoratively. That gives the core twenty-odd letters a very high hit rate on Instagram, X, Discord, Telegram, Reddit and Notion. Failures cluster at the edges: older Samsung One UI builds, in-app browsers with restricted font stacks, and any PDF where the font has been subsetted to the characters actually used will drop U+A730, U+A731 and U+A7AF first. The subtler problem is optical, not missing glyphs. Because the letters were encoded across three blocks over fifteen years, and each font vendor drew them to match different reference designs, stroke weights and heights drift within a single word, ɪ and ᴏ often sit a fraction taller than ꜱ. Real small caps are drawn as one coordinated set and optically corrected against the lowercase; this alphabet was assembled, not designed, and close up it shows.
Small capitals read as emphasis rather than volume, and that is the whole reason to use them. ALL CAPS carries a shouting convention that hardened in the 1980s on bulletin boards and never softened; a line set in small caps has the same even colour and horizontal rhythm as capitals but sits at lowercase height, so it reads as a label. Typographers have used exactly that quality for centuries in running heads, the first line of a chapter, and legal defined terms. In plain-text fields where bold and italic are simply unavailable, that distinction is worth something. A profile bio can mark a role line, a section break, or a set of contact labels without resorting to emoji bullets or capitals. Discord channel topics and Notion callouts use it the same way. Keep it to short spans, three or four words, a name, a category label. A full sentence in small caps loses the height variation that lets you recognise word shapes at a glance, and reading speed drops for the same reason it drops in ALL CAPS.
Screen readers handle these characters unpredictably, and that is the strongest argument against using them in anything load-bearing. Because ɢ and ᴋ are phonetic symbols rather than letters, a screen reader consults its character table: NVDA and VoiceOver may announce "Latin letter small capital G", read the IPA sound name, spell the string letter by letter at dictionary-entry pace, or say nothing at all. Behaviour varies by reader, by voice, by language pack and by version, so you cannot test once and assume. A bio line that takes you two seconds to read can take a screen-reader user thirty, or vanish. Search is the second cost. Platform search, browser find-in-page and site indexers all match on the actual codepoints, so a bio written in small caps will not surface for someone typing the plain word, ᴘʜᴏᴛᴏɢʀᴀᴘʜᴇʀ and photographer share nothing a matcher can see. Autocomplete, translation and text-to-speech fail the same way. The honest guidance is to style the decorative line and leave anything you want found, read aloud, or understood in plain characters.
Every small capital is a single codepoint in the Basic Multilingual Plane, which matters more than it sounds. Platforms that impose a character cap almost always count UTF-16 code units, and a BMP character is one unit, so ᴀʙᴏᴜᴛ costs exactly the same five characters as "about". That puts small caps among the cheapest styles available: Instagram's 150-character bio, X's 160, WhatsApp's 139-character About and Facebook's 101 all treat them as one-for-one. Compare that with strikethrough, underline or glitch styles, which stack combining marks on top of each letter, one measured 120 extra characters on a 48-character string, enough to overflow every field listed. The blocking constraint is elsewhere. Handles and usernames (an Instagram or X username, a Facebook name field) are validated against a restricted ASCII set and reject these characters outright, usually with a generic error. Some platforms also normalise unusual characters in display names during review. Display names, bios, captions, channel topics and post bodies accept them; anything that forms part of a URL or a login does not.
This is not a substitute for real small caps in a document. If you are working in Word, Google Docs, InDesign or CSS, use the typographic feature instead: Word's Font dialog has a Small caps checkbox, InDesign exposes OpenType smcp directly, and CSS gives you font-variant-caps: small-caps. Those keep the underlying text as ordinary letters, so search, spell-check and screen readers all still work, and the glyphs are the ones the type designer drew. Pasting output from here into a document defeats every one of those benefits. The tool also has no digits and no accents: Unicode encodes no small-capital numerals, and there are no precomposed small capitals for é, ñ or ü, so those pass through unchanged and break the line's rhythm. Names in languages that need them will look half-converted. Nor is it a font you can install or select. There is no download, no licence, nothing to embed, the styling lives in the characters themselves, which is exactly why it survives a paste into a field that has no font control at all.
Unicode has no small capital x, and the small capital q (U+A7AF) only arrived in 2018, so most fonts lack it. This tool uses ǫ for q and a plain lowercase x, which sit close to small-cap height but not on it. Words with q or x always look uneven.
Yes. Each small capital is a single Basic Multilingual Plane codepoint, so it consumes one of Instagram's 150 characters exactly like a plain letter. Unlike strikethrough or glitch styles, which stack combining marks, small caps never inflate your character count.
No. Instagram, X, TikTok and Facebook validate usernames against a restricted ASCII set and reject these characters, usually with an unhelpful generic error. Display names, bios, captions and channel topics accept them; anything that becomes part of a URL or login does not.
Unpredictably. Because these are phonetic symbols, NVDA or VoiceOver may say "Latin letter small capital G", spell the string out slowly, or skip it entirely. Behaviour changes with the reader, voice and language pack, so keep essential information in plain characters.
No. Word applies the OpenType smcp feature, which swaps glyphs while the underlying text stays ordinary letters, searchable, spell-checked, readable by assistive software. This tool swaps the characters themselves, which is the only approach that survives a paste into a plain-text field.
Not by searching those words. Platform search and find-in-page match raw codepoints, so ᴘʜᴏᴛᴏɢʀᴀᴘʜᴇʀ does not match photographer. Style a heading or a decorative line, and keep the keywords you want discovered in plain text somewhere on the profile as well.
The glyphs come from three Unicode blocks encoded between 1991 and 2006, drawn by font vendors to different reference designs. Real small caps are cut as one coordinated set and optically corrected; this alphabet was assembled from phonetic symbols, so weights and heights drift.
No. Unicode encodes no small-capital digits and no precomposed small capitals for é, ñ, ü or similar. Those characters pass through unchanged, so a name or date mixed into styled text will look half-converted. Keep such lines plain, or accept the inconsistency.