Which Base32 alphabet does this use?

It uses the standard RFC 4648 alphabet: the uppercase letters A–Z and the digits 2–7, with = for padding. This is the same encoding used by TOTP authenticator secrets.

How does Base32 encoding work?

It reads the input as a stream of bits and regroups them into 5-bit chunks (2^5 = 32), mapping each chunk to one alphabet character. Every 40 bits of input become 8 characters, and the output is padded with = signs to a multiple of 8.

Why is Base32 used instead of Base64?

Base32 is case-insensitive and avoids easily confused characters, so it survives being typed by hand, spoken aloud or stored in case-insensitive systems like DNS. It is less compact than Base64 but more robust.

Why are 0, 1, 8 and 9 missing from the alphabet?

The RFC 4648 alphabet deliberately omits digits that look like letters (0/O, 1/I/l) to reduce transcription errors. It keeps only 2–7 alongside A–Z to make 32 unambiguous symbols.

Is decoding case-sensitive?

No. The decoder upper-cases your input first, so both lowercase and uppercase Base32 decode correctly. Whitespace and trailing = padding are ignored.

Why is my Base32 longer than the original text?

Base32 packs 5 bits per character versus 8 bits per byte of input, so encoded output is about 60% larger than the raw text. That is the trade-off for being case-insensitive and DNS-safe.

No — encoding and decoding happen entirely in your browser with no network requests.

Base32 Encoder & Decoder

Base32 encoder and decoder (RFC 4648)

Base32 encodes binary data using 32 case-insensitive characters — the uppercase letters A–Z and the digits 2–7 — defined in RFC 4648. Because it avoids ambiguous characters and is case-insensitive, it is ideal for values that may be typed by hand or read aloud, such as TOTP two-factor secrets and DNS-safe identifiers.

How it works

To encode, the tool converts your text to UTF-8 bytes, then treats those bytes as one long bit stream and slices it into 5-bit groups. Each group (0–31) maps to one alphabet character. Since 5 and 8 do not divide evenly, the output is padded with = until its length is a multiple of 8. Decoding reverses the process: each character contributes 5 bits, which are reassembled into 8-bit bytes and decoded as UTF-8. The decoder upper-cases input and ignores whitespace and padding.

Example

Encoding the text Gera:

Gera → bytes 47 65 72 61 → I5SXEYI= (after padding)

Input	Base32
`A`	`IE======`
`Hi`	`JBUQ====`
`Gera`	`I5SXEYI=`

Everything runs in your browser — your data is never uploaded.