Base64 Encode & Decode

What Base64 is, and why it exists

Base64 represents arbitrary bytes using only 64 safe characters: A–Z, a–z, 0–9, + and /. Every three bytes of input become four characters of output, and when the input length is not a multiple of three, one or two = signs pad the end.

It exists because many channels were built for text, not binary data. Email (MIME), JSON, XML, HTML attributes, HTTP headers and environment variables can all carry Base64 safely where raw bytes would be mangled. That is why you meet it in email attachments, data: URIs, Basic authentication headers, JSON Web Tokens and API payloads that embed files.

Base64 is an encoding, not encryption. Anyone can decode it, so it hides nothing; if a value needs to stay secret, encrypt it. If you need to check that data has not changed, a checksum from the hash generator is the right tool.

Standard Base64, Base64URL and line wrapping

The decoder accepts all of these: either alphabet, with or without padding, with or without line breaks and spaces. A full data:…;base64, URI can be pasted as it is.

Standard (RFC 4648)

Uses + and / and pads with =. This is what MIME, data URIs, most libraries and the base64 command-line tool produce.

URL-safe, Base64URL

Replaces + with - and / with _ so the result can sit in a URL or a filename without percent-encoding. JWTs use it with the padding removed, which is why the padding option exists.

MIME line wrapping

Email requires lines of at most 76 characters, so MIME encoders insert a line break every 76 characters. Turn it on when pasting into an email source or a PEM-like block; leave it off for JSON, URLs and code.

Why text has to become UTF-8 first

Base64 encodes bytes, not characters, so text has to be turned into bytes before it can be encoded. The browser's built-in btoa() only accepts characters up to U+00FF and throws on anything else, which is why naive encoders fail on "€", "日本" or an emoji. This tool encodes text as UTF-8, the encoding nearly every system expects, so é becomes the two bytes C3 A9 and decodes back to é on any platform.

When decoding, the bytes are checked to be valid UTF-8. If they are not, the Base64 almost certainly holds a binary file rather than text, and the tool says so — and tells you which kind of file when it recognises the signature — instead of showing a screen of garbage characters.

Files: size, signatures and large inputs

Base64 makes data about a third larger: four characters for every three bytes, plus line breaks if you wrap. A 3 MB image becomes roughly 4 MB of text. The file view shows both sizes before you copy anything, which is worth a look before embedding a large file in a stylesheet or a JSON document.

Decoding a Base64 string to a file works the other way round. The first bytes of most file formats are a fixed signature — 89 50 4E 47 for PNG, %PDF for PDF, PK for ZIP (and so DOCX and XLSX) — so the tool reads them to name the download correctly and to preview images. Large files are encoded in chunks with a progress bar, so the page stays responsive, and very long results are previewed in part while copy and download always give you the whole thing.

For images specifically, the image to Base64 converter also writes ready-to-paste HTML, CSS and Markdown snippets.