Base32 Text Encoder Decoder

Encode UTF-8 text as RFC 4648 Base32 or strictly decode it back to text, locally in your browser.

RFC 4648 uppercase padded Base32. Text input limit: 50,000 UTF-16 code units; encoded input limit: 240,000 characters. Processing stays in your browser. Decoding rejects whitespace; pasted line endings may normalize to LF.

Encode UTF-8 text as canonical Base32

Paste text, leave decoding unchecked and choose Process input. Encoding converts Unicode text into UTF-8 bytes, then uses the RFC 4648 alphabet ABCDEFGHIJKLMNOPQRSTUVWXYZ234567. Output is uppercase with = padding to a multiple of eight characters. This is ordinary Base32, not Base32hex, Crockford Base32 or an authenticator secret generator.

Reproduce a short example

Load example (replaces input and mode) inserts hello. Processing produces NBSWY3DP, representing five UTF-8 bytes. Paste that result, enable Decode bytes to UTF-8 text, and process again to recover hello. For a shorter example, f becomes MY======.

Why decoding can fail

This tool intentionally requires canonical uppercase spelling, required padding and zero unused trailing bits. It rejects lowercase, spaces, line breaks, interior padding and omitted padding. For example, MZ====== is rejected even though its full eight data bits resemble f: the unused bits are nonzero. Do not guess at missing characters when handling identifiers supplied by another system.

The alphabet and padding follow RFC 4648. Applications can define other acceptance policies; their compatibility does not mean this strict text interface accepts every variant.

Limits and downloads

Text input must contain 1-50,000 UTF-16 code units; encoded input supports up to 240,000 characters. The decoder limit accommodates the largest supported UTF-8 encoding result. Empty input is treated as a missing entry, not an empty-byte conversion. No Unicode normalization is applied.

UTF-8 validation and export behavior

Decoding requires valid UTF-8: isolated continuation bytes, overlong sequences and encoded surrogate values fail instead of becoming replacement characters. A leading UTF-8 BOM is preserved as U+FEFF. Input containing an unpaired surrogate is rejected. A text encoding changes representation; it does not encrypt the text or prove its origin.

The TXT download contains the exact result encoded as UTF-8, without an added newline. Invisible control characters remain in the file. Browser textarea input can normalize CRLF or CR to LF, so this is not a byte-preserving file converter. Editing the input or mode removes the previous result and download. Copy can depend on browser permissions; downloading is an alternative.

Use UTF-8 Byte Inspector to see byte offsets and values, or Hex Text Encoder Decoder when you need a byte-pair representation. All conversion happens in this browser; the tool does not upload your text.