
Last checked by the TextEditor.com team
How to find a character by name or code
Type a word from the character's name, like "star", "degree" or "check". Results show up while you type, grouped by Unicode block. We also added everyday words that are not in the official names, so "tick" finds ✓ and "copyright" finds ©.
- Search. Type a name or a code, or paste the character itself.
- Inspect. Right click a tile, long press it, or press I to see its name, block and every code.
- Copy. Click a tile to copy the character, or use the Copy button next to the code you need.
The search box understands codes in all the usual forms. Click an example to try it.
| Form | Example | Where you see it |
|---|---|---|
| Code point | Unicode charts | |
| Hex number | Programming | |
| HTML entity, decimal | Web pages | |
| HTML entity, hex | Web pages | |
| HTML entity, named | Web pages | |
| JavaScript and JSON | Code | |
| CSS | Style sheets | |
| URL encoding | Web addresses | |
| The character | Anywhere |
What is this character? Paste it to find out
Sometimes text looks fine but acts strange. A search finds nothing, code won't run, or two names that look the same don't match. The cause is often a character you can't see, or one that only looks like another.
Press "Paste text to inspect" in the tool, or try one of these examples. You get one row per character, with its name and code.
Word and Google Docs turn straight quotes into curly ones (“ ” ‘ ’) while you type. That looks nice in a letter, but it breaks code and spreadsheet formulas. A no-break space (U+00A0) looks like a normal space, and it often comes along when you copy code from a web page. Many programming languages stop with an error when they meet one.
A zero width space (U+200B) has no width at all. Two usernames can look the same while one of them has it inside. The paste box marks such characters, and its "Copy without hidden characters" button gives you clean text. Need a blank for an empty name? The invisible character page lists every blank and where it works.
How Unicode is organized: planes and blocks
Unicode gives every character a number, called a code point. It is written as U+ and a hex number, like U+0041 for A. There is room for 1,114,112 code points, from U+0000 to U+10FFFF.
The space is cut into 17 planes of 65,536 code points each. Plane 0 is the Basic Multilingual Plane. It holds the letters of almost all modern languages and most common symbols. Most emoji live in plane 1. Planes 2 and 3 hold rare Chinese characters, and planes 4 to 13 are still empty.
Each plane is split into blocks: named ranges like Arrows (U+2190 to U+21FF). Pick a plane below to see its blocks, drawn wider when they are longer. Click a block to list its characters in the picker. Blocks with their own page on this site have an orange line.
Plane 0, the Basic Multilingual Plane: 55,747 characters.
Planes 15 and 16 are for private use: their code points have no meaning in Unicode, and each font or app can use them as it likes.
How many characters are in Unicode?
Unicode 18.0 came out in September 2026 and added 13,007 characters. That leaves most of the space free for later. Our count works the same way as Unicode's own: it leaves out control codes, private use code points and surrogates.
Chinese characters, also used in Japanese and Korean, are the biggest group by far. 101,985 of them are "CJK unified ideographs". Their names are just the code point, like CJK unified ideograph-4E00, so the search box finds them by code, not by meaning.
See what changed in each release on Unicode's Unicode 18.0 page. The block map and the counts on this page are built from Unicode's own data files.
HTML entities, CSS escapes and JavaScript escapes
On a modern web page saved as UTF-8, you can paste ★ or é straight into the HTML. You don't need a code for them.
Codes still help in a few places. In HTML, write <, > and & when you mean the signs <, > and &, or the browser may read them as code. An invisible character like the no-break space is easier to spot as than as a blank. In CSS, the content property takes escapes like \2605. In JavaScript and JSON, \u2605 is plain ASCII, so it survives files with the wrong encoding.
Café ★ 5 < 6Café ★ 5 < 6Caf\E9 \2605 5 < 6Caf\u00E9 \u2605 5 < 6Caf%C3%A9%20%E2%98%85%205%20%3C%206Characters above U+FFFF, like most emoji, take two code units in JSON and in older JavaScript: 😀 is \uD83D\uDE00. Newer JavaScript also accepts \u{1F600}. The inspector shows both forms.
How to type special characters on Windows, Mac and phones
Every system has a built in way to type symbols. None of them covers all of Unicode, which is why people copy from pages like this one. To type with an on-screen keyboard that has many alphabets, try the online Unicode keyboard.
| Method | System | How | Limits |
|---|---|---|---|
| Emoji panel | Windows 10 and 11 | Press the Windows key and the period key. The Ω tab has symbols. | A short list of symbols |
| Character Map | Windows | Search for Character Map in the Start menu. | Shows one font at a time |
| Alt codes | Windows | Hold Alt and type 0169 on the number pad to get ©. | Needs a number pad. Only a few hundred characters |
| Code, then Alt+X | Microsoft Word | Type 2605, then press Alt+X to get ★. | Works in Word, not in most other apps |
| Character Viewer | Mac | Press Control, Command and Space, then search by name. | Opens small at first. Click the button in its corner for the full viewer |
| Option key | Mac | Option+G types ©, Option+8 types •. | A set of common symbols only |
| Press and hold | Mac, iPhone, Android | Hold a letter key, like e, to pick é or è. | Accents and a few signs only |
| Ctrl+Shift+U | Linux with GNOME | Press Ctrl+Shift+U, type 2605, then press Space. | Depends on the input method |
The inspector lists the Windows Alt code and the Mac Option keys for each character that has them. If you only need the basic codes, the ASCII table with all 128 codes is quicker. For ticks and list bullets, see the page of check marks and bullets.
Why a character shows as a box
Unicode decides what a character is. A font decides how it looks. When no font on a device has a drawing for a character, you see an empty box, a box with a question mark, or nothing at all. People call that box "tofu".
Your system tries more than one font. If the font in use has no drawing, it looks for another font that has one. This is called font fallback, and it is why one symbol in a line can look a bit different from the rest.
The fix is a font that covers more of Unicode. Google's free Noto fonts aim to cover all of it, and the name comes from "no tofu". The test below draws six characters from six blocks with the fonts on your device.
From left: Miscellaneous Symbols, Mathematical Alphanumeric Symbols, Cyrillic Extended-B, Egyptian Hieroglyphs, Symbols for Legacy Computing and Miscellaneous Symbols Supplement, which is new in Unicode 18.0.
The inspector runs the same test on any character. If your device can't draw it, the inspector says so. The character still copies, and it shows on any device that has a fitting font.
A short history: from ASCII to Unicode
Before Unicode, each computer maker and country had its own character sets. The same number could mean é on one machine and Θ on another, so text often turned into nonsense when it moved between systems.
1963
The first ASCII standard: 128 codes for the English alphabet, digits, punctuation and control codes.
1981
The IBM PC uses code page 437. It adds accented letters, some Greek letters and the box drawing pieces of DOS programs.
1985
Windows 1.0 uses Windows-1252 for Western European languages. Its Alt+0 codes still work today.
1991
Unicode 1.0 comes out in October. The goal: one number for every character in every language.
1996
Unicode 2.0 adds surrogates. They make room for more than 65,536 characters, up to 1,114,112 code points.
2010
Unicode 6.0 adds emoji from Japanese phones. See the emoji picker for all of them.
2025
Unicode 17.0 reaches 159,801 characters.
2026
Unicode 18.0 adds 13,007 characters, for 172,808 in all.
Today almost every web page uses UTF-8, a way to store Unicode text that keeps the old ASCII codes as they were. That is why plain English text from 1963 still reads fine. You can also turn text into fancy letters made from Unicode, and the page of box drawing characters has the lines and blocks from code page 437.
More symbol tools
- Invisible character[ㅤ]A blank to copy for empty names and messages.
- Search every emoji😀 🐶 🍕 ⚽One search box for all emoji and flags.
- Font converter𝐀 𝒜 𝔄 𝔸 Ⓐ 𝙰Your text in more than 60 letter styles.
- Encode and decodeHi → SGk=Base64, URL codes, HTML entities and hex.
- Text frames█▓▒▒░░░ Hi ░░░▒▒▓█Boxes, borders and one line frames around text.