Do I need to clean up my hex dump first?

No. Spaces, commas, colons and `0x` prefixes are ignored, and capitals read the same as lowercase. Paste what your debugger, your log or your hex editor gave you.

Why does it refuse an odd number of digits?

Because a byte is two digits, and there is no way to know which end is missing. Guessing would silently produce a different byte than the one you have — the tool says so instead, and tells you how many digits it counted.

Why is there a replacement character in the output?

Because the bytes are not valid UTF-8 — often because the dump is truncated mid-character. An accented letter is two bytes and an emoji is four; cut one in half and no text can be recovered from the remainder.

Hex to Text — Paste a Dump However You Got It

Decode hexadecimal bytes back to UTF-8 text. Spaces, commas, colons, 0x prefixes and capitals are all accepted, because dumps come in every shape.

Instant Private Zero cookies

What this tool does

The hexadecimal digits are read two at a time into bytes, and those bytes are decoded as UTF-8 text.

It accepts dumps as they come

Hex arrives in as many shapes as there are tools that print it, so the separators are simply ignored:

48 65 6c 6c 6f      a hex editor
48,65,6c,6c,6f      a CSV export
48:65:6c:6c:6f      a MAC address, a certificate fingerprint
0x48 0x65 0x6c      C source, a debugger
48656c6c6f          a compact dump

All five decode to the same thing. Capitals read the same as lowercase.

What it refuses, and why

An odd number of digits. A byte is two digits; with an odd count, one end is missing and nothing says which. Padding a zero on the left or the right produces a different byte, so the tool reports the count instead of choosing for you.

A character that is not a hexadecimal digit. The message names it and gives its position in what you typed, not in the text after separators were stripped — the two differ, and pointing at the wrong character would be worse than saying nothing.

Replacement characters in the output

These mean the bytes are not valid UTF-8, and the usual cause is a truncated dump. An accented letter takes two bytes, a CJK character three, an emoji four. Cut one of them in half and the remainder cannot be read as text — by this tool or any other.

If you expected readable text, check whether the dump covers whole characters rather than a fixed number of bytes.

Private by design

Everything runs locally in your browser with JavaScript. Your data is never uploaded, which makes the tool safe for sensitive content, and it keeps working offline.

Frequently asked questions

Do I need to clean up my hex dump first?: No. Spaces, commas, colons and `0x` prefixes are ignored, and capitals read the same as lowercase. Paste what your debugger, your log or your hex editor gave you.
Why does it refuse an odd number of digits?: Because a byte is two digits, and there is no way to know which end is missing. Guessing would silently produce a different byte than the one you have — the tool says so instead, and tells you how many digits it counted.
Why is there a replacement character in the output?: Because the bytes are not valid UTF-8 — often because the dump is truncated mid-character. An accented letter is two bytes and an emoji is four; cut one in half and no text can be recovered from the remainder.