🔬 Unicode Converter
Convert text to various Unicode representations and decode them back. Handles emoji and surrogate pairs in most formats.
Last updated: November 8, 2026 · By the BoltQuickTools team
Decode Unicode Back to Text
Paste any encoded format (U+, \u, &#, etc.) and it will be auto-detected and decoded.
See exactly which characters a string contains
A pasted value fails a comparison, a CSV shows odd symbols, or you need an emoji written as an escape in source code. Those problems come down to the actual code points and bytes in the text, which an editor does not show. This page lays them out in the notations developers use.
Encoding, decoding and inspecting
Type into Input Text and eight cards fill in as you type: Unicode Code Points, JavaScript Escape, HTML Decimal, HTML Hex, CSS Escape, Python Escape, UTF-8 Hex Bytes and UTF-16 Hex. Each has its own Copy button. The buttons Hello World, Emoji Test and CJK Sample load test strings.
To go the other way, paste into the box under Decode Unicode Back to Text. It decodes as you type or when you press Decode, and it names the format it detected. The Character Inspector table lists each character with its code point, bytes and a category.
Example: H, e with an accent, and an emoji
The three-character input Hé😀 produces:
| Card | Output |
|---|---|
| Unicode Code Points | U+0048 U+00E9 U+1F600 |
| JavaScript Escape | \u0048\u00E9\u{1F600} |
| HTML Decimal | Hé😀 |
| HTML Hex | Hé😀 |
| CSS Escape | \0048\00E9\1F600 |
| Python Escape | \u0048\u00e9\U0001f600 |
| UTF-8 Hex Bytes | 48 C3 A9 F0 9F 98 80 |
| UTF-16 Hex | 0048 00E9 D83D DE00 |
The bytes show why string lengths disagree between systems: that is 3 code points, 4 UTF-16 units and 7 UTF-8 bytes.
Where the output needs care
- The JavaScript form is not valid JSON for emoji.
\u{1F600}works in modern JavaScript, but JSON needs the surrogate pair\uD83D\uDE00. Take those two values from the UTF-16 Hex card. - Every character is escaped, including plain ASCII letters, so the output is much longer than the input.
- CSS escapes have no trailing space. If you paste one directly before a letter or digit from a to f or 0 to 9, CSS reads that next character as part of the escape. Add a space after the escape in that case.
- Grouped characters are shown in pieces. A heart emoji with a colour selector appears as two rows, and flags and family emoji split into their parts.
- The inspector is partial. It stops after 500 characters. Names cover ASCII, CJK ideographs and a short list of common symbols, so many rows are blank (
éhas no name), and Category is a rough range-based guess:$is labelled punctuation although Unicode classes it as a currency symbol.
How decoding picks a format
The checks run in a fixed order and the first match wins, which has some consequences:
- Python output containing emoji only half decodes. The mix of
\uand\Uabove is detected as JavaScript, givingHé\U0001f600. A string that uses only\Uescapes decodes correctly. - Short words look like hex.
de adis decoded as two UTF-8 bytes. Without spaces,4865is read as a single UTF-16 unit. - The U+ form needs four to six digits.
U+48is rejected with a message listing the accepted forms, andU+110000shows an invalid code point error. - Escapes can sit inside other text.
price: \u20AC5becomesprice: €5.
Questions about text encoding
Why do my UTF-8 bytes decode to replacement symbols?
The hex decoder expects complete UTF-8 sequences. If a multi-byte character has been cut off, the browser's decoder puts a replacement character in its place.
Does this cover named entities like é?
No. It handles numeric entities only. The HTML entity tool handles named ones, and percent-encoded text such as %C3%A9 belongs in the URL encoder.
Where can I read more about encodings?
The character encoding article explains how code points, UTF-8 and UTF-16 relate.