Base64 Encoder / Decoder

Encode or decode text and files as Base64 or Base64URL. 100 % local processing in your browser: no data is sent to a server.

Variant

Standard

Base64URL

Text

Source

Base64

Result

Learn more about Base64

What is this tool?

Base64 is a way to write binary data as plain ASCII text. Groups of 3 bytes (24 bits) are split into 4 groups of 6 bits, and each group becomes one character from a 64-character alphabet. The result can travel through systems that only accept text, such as email, JSON, XML or a URL.

This tool encodes text or a file to Base64, and decodes Base64 back to text, as you type. The area you edit is the source and the other one shows the result. You can use the standard alphabet or the URL-safe variant, Base64URL.

Typical use cases

Embedding a small image or file in JSON, CSS or an HTML data URL.
Decoding a value copied from an API, a log or an email attachment.
Producing a Base64URL string that can be placed in a URL path or a file name.
Checking the raw bytes of a PEM block after removing its BEGIN and END lines.

How to use it

Paste text into the left area, or Base64 into the right area. The other area updates on its own.
The area you are editing is marked as the source. Editing the result turns it into the new source.
Choose Standard or Base64URL. The Base64 area is regenerated from the current source.
To encode a file, click Encode a file. The bytes appear as Base64. A binary file is not shown as text.
Use Copy above an area, or Download to save the raw bytes.

Encoding and decoding run entirely in your browser. The text and the file stay on your machine: nothing is uploaded.

Before / after examples

RFC 4648 test vector (standard)

Before

foobar

After

Zm9vYmFy

Text, standard alphabet

Before

Hello, DTK!

After

SGVsbG8sIERUSyE=

The same bytes as Base64URL

Before

Hello, DTK!

After

SGVsbG8sIERUSyE

Hello, DTK! is 11 bytes, so the standard form ends with one = of padding. Base64URL keeps the same data and drops that padding. The accented word café is encoded from its UTF-8 bytes as Y2Fmw6k= .

Common mistakes

Base64 looks like a scrambled secret, but the transformation is public and reversible. Most mistakes come from the alphabet, the character encoding, or extra text wrapped around the payload.

café encoded as Latin-1 instead of UTF-8

Invalid

Y2Fm6Q==

Valid

Y2Fmw6k=

Standard Base64 inside a URL

Invalid

Pj4/

Valid

Pj4_

Data URL prefix left in the payload

Invalid

data:text/plain;base64,SGVsbG8=

Valid

SGVsbG8=

Base64 is not encryption. Anyone can decode it, with no key. Do not use it to hide a password, a token or personal data.

Good to know

Text is encoded as UTF-8 before Base64 is applied, and decoding expects UTF-8. Bytes that are not valid UTF-8, such as a PNG or a PDF, are not shown as text: use Download to save them.
On decode, whitespace is ignored and missing = padding is restored. Both alphabets are accepted, so a standard value still decodes while Base64URL is selected.
On encode, Standard keeps + , / and the = padding of RFC 4648. Base64URL uses - and _ and omits the padding. Changing the variant re-encodes from the source.
If the source is the Base64 area and you change the variant, the bytes are encoded again and the text area becomes the source. An invalid Base64 value is left unchanged.
Files larger than 4 MB (4 × 1024 × 1024 bytes) are refused, so the page stays responsive.
A share link stores the source text or the Base64 string, the variant, and the file name when the source is a file. The other area is computed when the link is opened. A large file produces a very long link.

Did you know?

The current specification is RFC 4648, published in October 2006 by Simon Josefsson. It obsoletes RFC 3548.
The name comes from the alphabet size: 2 to the power 6 is 64. Each character carries 6 bits, so every 3 bytes become 4 characters, about one third larger.
MIME, defined in RFC 2045 (November 1996), uses Base64 and limits each line to 76 characters so the text survives old mail gateways. This tool ignores those line breaks when decoding.
One of the official test vectors in RFC 4648 encodes the word foobar as Zm9vYmFy .
JSON Web Signature (RFC 7515, May 2015) requires Base64URL, and it omits the = padding, for the protected header and the payload.
A PEM certificate, as described by RFC 7468 (April 2015), is a Base64 body wrapped with -----BEGIN and -----END lines. Those lines are not part of the Base64 payload.

Frequently asked questions

Is my data sent to a server?: No. Encoding, decoding and the share link are handled in your browser. The content is neither stored nor transmitted. The share link keeps the data in the URL fragment, which is not sent to the server.
Does Base64 protect a secret?: No. Base64 is an encoding, not a cipher. Decoding does not require a password. Use real encryption if the content must stay confidential.
What is the difference between Standard and Base64URL?: The data is the same. Standard uses + and / , and pads with = so the length is a multiple of 4. Base64URL replaces those two characters with - and _ , and this tool omits the padding, which is the form used in URLs and in JSON Web Tokens.
Why does my file not appear as text?: Only valid UTF-8 can be shown in the text area. An image, a PDF or any other binary file is encoded to Base64, and Download saves those exact bytes. The original file name is kept for that download until you edit the text.
Why is the result longer than the input?: Four characters represent three bytes, so the text grows by about 33%. One or two = characters may be added in the standard variant so the length is a multiple of 4. They disappear in Base64URL.