Base32 / Base58 / Base62 / Base85 / Hex Encoder & Decoder
Pick a format and type or paste text to convert it instantly. Switch to "Decode" to turn an encoded string back into text. For Base64, use the Base64 Encoder/Decoder; for number systems, try the Base Converter.
🔒 Runs in your browser · Your data is never sent to a server
Notes
- These are encodings, not encryption. Anyone can reverse them, so they cannot protect passwords or other secrets.
- Text is handled as UTF-8 bytes, so Japanese characters and emoji work. Data made from another encoding may fail to decode.
- Base32 follows RFC 4648, Base58 uses the Bitcoin alphabet (no 0, O, I or l), and Base85 is Ascii85 (Adobe). Results differ from Z85 or other Base58 alphabets.
- Base58 and Base62 convert the data as one huge number, so inputs of tens of thousands of characters can be slow.
Glossary
Base32
Represents binary data with 32 characters (A-Z and 2-7). Case mix-ups are rare, and it is used for things like TOTP secret keys.
Base58
Uses 58 characters, leaving out look-alikes (0, O, I, l) and symbols. Best known from Bitcoin addresses.
Base62
Uses only digits and upper/lowercase letters (0-9, A-Z, a-z). With no symbols, it suits short URL IDs.
Base85 (Ascii85)
Encodes 4 bytes as 5 characters, so output is shorter than Base64 (about 1.25x). Used in PostScript and PDF.
FAQ
When should I use Base32, Base58, Base62 or Base85?
Base32 is case-insensitive and easy to read aloud, so it is used for TOTP secrets. Base58 drops look-alike characters, which is why Bitcoin addresses use it. Base62 has no symbols, making it handy for short-URL IDs. Base85 gives shorter output than Base64.
Why does decoding say a character is not valid?
The input contains a character that is not in the alphabet of the selected format. Base58 has no 0, O, I or l, and Base32 has no 0, 1, 8 or 9. Check that the format matches and that no extra characters were copied. Spaces and line breaks are ignored.
Can the Base58 result differ from other tools?
This tool uses the Bitcoin alphabet (123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz) and writes each leading 0x00 byte as "1". Results will not match Base58 variants with a different alphabet, such as Ripple, or Base58Check, which adds a checksum.
Can it decode Base85 wrapped in "<~" and "~>"?
Yes. The "<~" and "~>" markers around Ascii85 data are removed automatically, and the "z" shortcut for four zero bytes is supported. Variants with a different alphabet, such as Z85, are not supported.
Share this page