Decode Base58 text without grabbing the wrong alphabet

Base58 exists because 0/O/I/l look alike — but three different alphabets use that name. This codec uses the Bitcoin one.

Tools used in this guide

The job

Bitcoin addresses, IPFS hashes and Solana keys all say 'Base58'. The Bitcoin alphabet and the Flickr alphabet differ in letter order — decoding with the wrong one yields garbage that still looks like bytes.

Paste and decode

Paste the Base58 string and decode. This tool uses the Bitcoin alphabet — a string written in the Flickr alphabet needs a different decoder. The decoded bytes come back as UTF-8 text; binary payloads that aren't valid UTF-8 show replacement characters.

  • Leading zeros matter: each leading '1' encodes a 0x00 byte and survives the round-trip.
  • Base58Check payloads include a 4-byte checksum trailer — this decodes the bytes; verifying the trailer is a separate step.

Encoding too

The same tool encodes: paste UTF-8 text and get the Base58 string. Input limit 1,000,000 characters either way.