Base64

Base64 is a way to represent binary data (bytes) as a safe, printable string made of letters, digits, and a few symbols. It is everywhere on the web - it powers data URIs in img tags, JWT tokens (eyJ…), and the bodies of many APIs. This tool converts text into Base64 and back.

How to use

1. Type or paste your text (or a Base64 string) into the left panel - the result updates as you type.
2. Encode / Decode - pick the direction with the segmented buttons:
   • Encode turns text into Base64. The text is encoded as UTF-8 bytes first, so accents, emojis, and non-Latin scripts all work.
   • Decode turns Base64 back into text. Text must be valid UTF-8; if the decoded bytes aren't, a clear error is shown instead of mojibake.
3. Swap - moves the current result into the input and flips the direction, so you can chain encode → decode → encode.
4. Copy - puts the current result on your clipboard.

Options

• URL-safe - uses - and _ instead of + and /, the variant that can appear safely in URLs and filenames (JWT uses this). Padding = is kept, so the two forms stay interchangeable.
• Ignore line breaks - removes whitespace and newlines from the input before decoding. Many Base64 outputs wrap at 76 characters, and pasting them back in needs this.

Notes and examples

• Hello encodes to SGVsbG8=. Try it, then hit Decode to get Hello back.
• A valid Base64 length is a multiple of 4, with = padding (0, 1, or 2 characters) filling the final group. The tool also accepts unpadded input like SGVsbG8.
• Invalid characters and misplaced padding produce an error that names the offending character, so you know exactly where the string is off.
• Base64 encodes bytes, not necessarily text. Decoding arbitrary bytes only succeeds when they happen to be valid UTF-8 - anything else is exactly what the error on non-UTF-8 behaviour is for.
• For reference: the empty string encodes to nothing (empty output), and each group of 3 input bytes becomes 4 Base64 characters.