Encode and decode Base64.

No. Encoding and decoding happen in this page, and nothing you paste or choose is sent anywhere.

Yes, for encoding, up to 1 MB. The tool shows the file's size and the Base64, ready to copy. Decoding works on text only, so it cannot rebuild a file.

The tool names the problem. The input may use characters outside the standard alphabet, such as the - and _ of URL-safe Base64, or be the wrong length, which usually means part of it was lost in copying. Valid Base64 that holds binary data, such as an image, is reported by its byte count.

Yes. Encoding, decoding and file mode need no account and cost nothing, and there is no limit on how often you use them.

Yes. Text is turned into UTF-8 bytes before it is encoded, and decoding reads UTF-8 back, so é, ß and emoji come out exactly as they went in. The browser's own btoa function throws an error on characters like these when it is used alone, which is why the tool converts to bytes first.

No. Base64 without its closing = signs decodes as long as the length is otherwise valid. An = in the middle of the text, or more than two at the end, counts as a character outside the alphabet.

Spaces and line breaks are removed before decoding, so Base64 copied out of an email, a key file or a wrapped log line decodes as one string.

When the box holds a result, switching between Text to Base64 and Base64 to text moves that result into the box. Encode some text, switch, and the Base64 is already waiting to be decoded. If the box is empty or shows an error, your input stays as it is.

Put a prefix in front of it that names the file type, such as data:image/png;base64, for a PNG, then paste the whole string where a URL goes in CSS or an img tag. The tool prints the Base64 alone, so the prefix is yours to add.

Base64 writes every 3 bytes as 4 characters, so the output is about a third longer than the input, plus up to two = signs of padding. File mode shows the file's size and the number of Base64 characters it became.

No. Base64 writes bytes as plain characters, and anyone can decode it. Use it to carry data through places that only accept text, not to hide it.

No. The text lives in this page while it is open. Nothing is saved in the browser or sent anywhere, so reloading the page starts with an empty box.

Any type. The file is read as raw bytes, so an image, a font, a PDF or a ZIP encodes the same way, as long as it is 1 MB or smaller.

No. It writes the standard alphabet, with + and / and = padding. For a value going into a URL, run the result through the URL encoder, or swap + for -, / for _ and drop the padding yourself.

No. The Base64 comes out as one unbroken line, which is what data URIs and most config values expect. The output box wraps it on screen only, and Copy output copies it without breaks.

No. Text is always encoded as UTF-8, and decoding reads UTF-8 too. Base64 made from Latin-1 text with accents is reported as binary data rather than text.

There is no set limit. Very long input takes longer to convert, since the result is worked out again on every change.

Base64 works in groups of 3 bytes. When the input does not fill the last group, one or two = signs pad the output to a multiple of 4 characters.

Not as it is. Remove everything up to and including the comma first, since the colon, the semicolon and the comma are outside the Base64 alphabet and the decoder reports them.