Short answer: Tap any numeral below to copy it. But read the next section first — for almost everything you’d write in English, you should type the ordinary letters I, V, X instead. The Unicode Standard says so itself.
Tap any numeral to copy it
Need other math characters? The full symbol keyboard has every operator, fraction and Greek letter with the same tap-to-copy.
Read this before you use them
Unicode contains dedicated Roman numeral characters: Ⅰ Ⅴ Ⅹ and the rest, each a single character with its own code point. They look identical to the Latin letters I, V and X — but they are not those letters.
The Unicode Standard’s own guidance is that for most purposes you should build Roman numerals out of ordinary Latin letters. The dedicated characters exist mainly for round-trip compatibility with older East Asian standards that encoded them separately, not because anyone writing English needs them.
What goes wrong when you use them anyway:
- Search breaks. Ctrl+F for
Vwill not find Ⅴ. Your reader can’t find the text, and neither can site search. - Sorting breaks. Ⅹ sorts nowhere near X, because it’s character 8553 rather than 88.
- Screen readers vary. Some announce “roman numeral ten,” some read the character name, some skip it. Plain
Xis announced predictably. - Fonts are patchy. Many fonts have no glyphs for these, so readers see boxes (why that happens).
- Copy-paste degrades. Paste Ⅻ into a form that strips non-ASCII and it vanishes entirely, rather than degrading to XII.
Type Chapter IV, not Chapter Ⅳ. Use the real characters when you’re working with Japanese, Korean or Chinese text that expects them, when you need the numeric property for parsing, or for the archaic forms below — which are meant for general use.
Full reference
| Symbol | Value | Unicode | HTML | Lowercase |
|---|---|---|---|---|
| Ⅰ | 1 | U+2160 | Ⅰ | ⅰ |
| Ⅱ | 2 | U+2161 | Ⅱ | ⅱ |
| Ⅲ | 3 | U+2162 | Ⅲ | ⅲ |
| Ⅳ | 4 | U+2163 | Ⅳ | ⅳ |
| Ⅴ | 5 | U+2164 | Ⅴ | ⅴ |
| Ⅵ | 6 | U+2165 | Ⅵ | ⅵ |
| Ⅶ | 7 | U+2166 | Ⅶ | ⅶ |
| Ⅷ | 8 | U+2167 | Ⅷ | ⅷ |
| Ⅸ | 9 | U+2168 | Ⅸ | ⅸ |
| Ⅹ | 10 | U+2169 | Ⅹ | ⅹ |
| Ⅺ | 11 | U+216A | Ⅺ | ⅺ |
| Ⅻ | 12 | U+216B | Ⅻ | ⅻ |
| Ⅼ | 50 | U+216C | Ⅼ | ⅼ |
| Ⅽ | 100 | U+216D | Ⅽ | ⅽ |
| Ⅾ | 500 | U+216E | Ⅾ | ⅾ |
| Ⅿ | 1000 | U+216F | Ⅿ | ⅿ |
The lowercase forms run U+2170–U+217F in the same order.
Ⅻ is a single character. Not three. It’s one code point that happens to render as XII — which is why length returns 1 for it in most programming languages, and why it breaks naive string handling.
The archaic forms — these ones are fine to use
Four characters sit just past the compatibility block, and they’re different in kind. Unicode intends these for general use with the Latin alphabet, not merely for East Asian compatibility:
| Symbol | Value | Unicode | HTML | Name |
|---|---|---|---|---|
| ↀ | 1,000 | U+2180 | ↀ | Roman numeral one thousand C D |
| ↁ | 5,000 | U+2181 | ↁ | Roman numeral five thousand |
| ↂ | 10,000 | U+2182 | ↂ | Roman numeral ten thousand |
| Ↄ | — | U+2183 | Ↄ | Roman numeral reversed one hundred |
Ↄ is exactly what it looks like — a reversed C. It combines with Ⅰ and Ⅽ in the apostrophus system the Romans used for large numbers, which is why ↀ carries the name “one thousand C D.”
Three more exist for specialist use: ↅ (six, late form), ↆ (fifty, early form), ↇ (fifty thousand) and ↈ (one hundred thousand).
How to type them
Windows
Alt codes don’t work. Every character here is above 255, so there’s no Alt + 0NNN route, and none appear in the legacy CP437 set either (why Alt codes fail).
- Word, Outlook, PowerPoint: type the hex then Alt + X.
2164+ Alt + X gives Ⅴ. - Anywhere else: Windows key + period, then the Ω symbol tab.
Mac
No Option shortcut exists for any of them. Open Character Viewer (Control + Command + Space) and search roman numeral. Or enable Unicode Hex Input under System Settings → Keyboard → Text Input → Edit → Add Input Source → Others, then hold Option and type the hex — Option + 2164 gives Ⅴ.
Chromebook and Linux
Press Ctrl + Shift + U, type 2164, press Enter or Space.
iPhone and Android
Not on either stock keyboard. Copy from this page and save a text replacement — full walkthrough in text replacement for symbols.
Numbered lists — let the software do it
If you want a Roman-numbered list, don’t paste characters. Every major tool generates the numbering for you, using ordinary letters, and renumbers automatically when you insert an item.
HTML and CSS:
html
<ol type="I">
<li>First</li>
<li>Second</li>
</ol>css
ol.outline { list-style-type: upper-roman; }Both produce I, II, III as normal Latin letters — searchable, sortable, screen-reader-friendly.
Word: Home → Numbering → choose the I, II, III format. Google Docs: Format → Bullets & numbering → Numbered list → Roman option. LaTeX: \Roman{section} or \roman{}, covered in LaTeX math symbols.
This is the single most common reason people look for these characters, and it’s the one case where the characters are definitively the wrong tool.
Where the real characters do belong
- Japanese, Chinese and Korean documents, where the compatibility characters are expected and render correctly in CJK fonts.
- Fixed-width or monospaced layouts where Ⅻ occupying one cell instead of three actually matters.
- Data with a numeric property — these characters carry a machine-readable numeric value (Ⅴ = 5), which parsers can read directly. Plain letters deliberately don’t, so that simple parsers don’t misread “II.A.1.b” in an outline as a number.
- Paleographic and historical typesetting, especially the apostrophus forms ↀ ↁ ↂ Ↄ.
Ⅴ or V — a quick test
Paste the character into a search box on the same page and press Enter. If a plain V doesn’t match it, you’re holding U+2164, not the letter. That’s the whole problem in one action.
Normalising text with Unicode NFKD folds Ⅴ back to V, ⅫC to XII — which is exactly how databases and search engines sometimes rescue this. Sometimes. Don’t rely on it.
Troubleshooting
| Problem | Fix |
|---|---|
| Numerals show as empty boxes ▯ | Font has no glyphs for U+2160–U+217F. Use plain Latin letters instead, or switch to a font with CJK coverage. |
| Ctrl+F can’t find them | Expected — Ⅴ is not V. Retype using ordinary letters. |
| They sort in the wrong order | Same cause. Sorting uses code point order, and these are at 8544+. |
| Pasting into a form strips them | The form filters non-ASCII. Use plain letters. |
| Alt codes produce nothing | All are above 255. Use hex + Alt + X, or the emoji panel. |
| A list renumbers wrongly | You pasted characters instead of using list numbering. Switch to a numbered list with the Roman format. |
| Ⅻ counts as one character in code | Correct — it is one code point. Normalise with NFKD to expand it to XII. |
| Screen reader skips them | Announcement is inconsistent across readers. Use plain letters in body text — see screen reader symbols. |
Recap
| Situation | Do this |
|---|---|
| Writing English text | Type I, V, X — ordinary letters |
| Numbered list | Use the tool’s Roman numbering format |
| CJK document | Use the real characters, U+2160+ |
| Windows, in Word | 2164 + Alt + X |
| Mac | Control + Command + Space, search roman numeral |
| Chromebook or Linux | Ctrl + Shift + U, 2164, Enter |
| A web page | Ⅴ, or <ol type="I"> for lists |
Frequently asked questions
Is the Roman numeral Ⅴ the same as the letter V?
No. Ⅴ is U+2164, a separate character in the Number Forms block. The letter V is U+0056. They look identical in most fonts, but searching for V will not match Ⅴ, and the two sort differently.
Should I use Unicode Roman numeral characters?
Usually not. The Unicode Standard recommends composing Roman numerals from ordinary Latin letters for most purposes, because the dedicated characters exist mainly for compatibility with older East Asian standards. Use plain I, V and X in English text.
How do I type Roman numerals on Windows?
In Word, Outlook or PowerPoint, type the hex code and press Alt + X — 2164 becomes Ⅴ. Alt codes do not work, because every one of these characters is above 255. Elsewhere, use the Windows key + period emoji panel.
Why is Ⅻ one character instead of three?
Unicode encoded the numerals through twelve as single characters, so Ⅻ is one code point with a numeric value of 12. This is why string length functions count it as one, and why normalising it expands it into XII.
How do I make a Roman numbered list?
Use your tool’s numbering feature rather than pasting characters — <ol type="I"> in HTML, list-style-type: upper-roman in CSS, or the Roman option in Word and Google Docs. These generate ordinary letters and renumber automatically.
What are ↀ ↁ and ↂ?
They are archaic Roman numerals for 1,000, 5,000 and 10,000, built from the apostrophus system that combined Ⅰ and Ↄ, the reversed C. Unlike the compatibility numerals, these are intended for general use with the Latin alphabet.
Written by Igor R. Tested on Windows 11, macOS, ChromeOS, Linux (IBus), iOS and Android. Last updated August 2026. Symbol code points follow the Unicode Standard.