base64 — Encode Binary Data with ASCII

Purpose:The base64 module contains functions for translating binary data into a subset of ASCII suitable for transmission using plaintext protocols.

The base64, base32, base16, and base85 encodings convert 8 bit bytes to values that fit inside the ASCII range of printable characters, trading more bits to represent the data for compatibility with systems that only support ASCII data, such as SMTP. The base values correspond to the length of the alphabet used in each encoding. There are also URL-safe variations of the original encodings that use slightly different alphabets.

Base 64 Encoding

This is a basic example of encoding some text.
import base64
import textwrap

# Load this source file and strip the header.
with open(__file__, 'r', encoding='utf-8') as input:
    raw =
    initial_data = raw.split('#end_pymotw_header')[1]

byte_string = initial_data.encode('utf-8')
encoded_data = base64.b64encode(byte_string)

num_initial = len(byte_string)

# There will never be more than 2 padding bytes.
padding = 3 - (num_initial % 3)

print('{} bytes before encoding'.format(num_initial))
print('Expect {} padding bytes'.format(padding))
print('{} bytes after encoding\n'.format(len(encoded_data)))

The input must be a byte string, so the unicode string is first encoded to UTF-8. The output shows the 185 bytes of the UTF-8 source expand to 248 bytes after being encoded.


There are no carriage returns in the encoded data produced by the library, but the output has been wrapped artificially to make it fit better on the page.

$ python3

184 bytes before encoding
Expect 2 padding bytes
248 bytes after encoding


Base 64 Decoding

b64decode() converts an encoded string back to the original form by taking four bytes and converting them to the original three, using a lookup table.
import base64

encoded_data = b'VGhpcyBpcyB0aGUgZGF0YSwgaW4gdGhlIGNsZWFyLg=='
decoded_data = base64.b64decode(encoded_data)
print('Encoded :', encoded_data)
print('Decoded :', decoded_data)

The encoding process looks at each sequence of 24 bits in the input (three bytes) and encodes those same 24 bits spread over four bytes in the output. The equal signs at the end of the output are padding inserted because the number of bits in the original string was not evenly divisible by 24, in this example.

$ python3

Encoded : b'VGhpcyBpcyB0aGUgZGF0YSwgaW4gdGhlIGNsZWFyLg=='
Decoded : b'This is the data, in the clear.'

The value returned from b64decode() is a byte string. If the contents are known to be text, the byte string can be converted to a unicode object. However, the point of using base 64 encoding is to be able to transmit binary data, and so it is not always safe to assume that the decoded value is text.

URL-safe Variations

Because the default base64 alphabet may use + and /, and those two characters are used in URLs, it is often necessary to use an alternate encoding with substitutes for those characters.
import base64

encodes_with_pluses = b'\xfb\xef'
encodes_with_slashes = b'\xff\xff'

for original in [encodes_with_pluses, encodes_with_slashes]:
    print('Original         :', repr(original))
    print('Standard encoding:',
    print('URL-safe encoding:',

The + is replaced with a -, and / is replaced with underscore (_). Otherwise, the alphabet is the same.

$ python3

Original         : b'\xfb\xef'
Standard encoding: b'++8='
URL-safe encoding: b'--8='

Original         : b'\xff\xff'
Standard encoding: b'//8='
URL-safe encoding: b'__8='

Other Encodings

Besides Base64, the module provides functions for working with Base85, Base32, and Base16 (hex) encoded data.
import base64

original_data = b'This is the data, in the clear.'
print('Original:', original_data)

encoded_data = base64.b32encode(original_data)
print('Encoded :', encoded_data)

decoded_data = base64.b32decode(encoded_data)
print('Decoded :', decoded_data)

The Base32 alphabet includes the 26 uppercase letters from the ASCII set and the digits 2 through 7.

$ python3

Original: b'This is the data, in the clear.'
Decoded : b'This is the data, in the clear.'

The Base16 functions work with the hexadecimal alphabet.
import base64

original_data = b'This is the data, in the clear.'
print('Original:', original_data)

encoded_data = base64.b16encode(original_data)
print('Encoded :', encoded_data)

decoded_data = base64.b16decode(encoded_data)
print('Decoded :', decoded_data)

Each time the number of encoding bits goes down, the output in the encoded format takes up more space.

$ python3

Original: b'This is the data, in the clear.'
Encoded : b'546869732069732074686520646174612C20696E207468652063
Decoded : b'This is the data, in the clear.'

The Base85 functions use an expanded alphabet that is more space-efficient than base 64.
import base64

original_data = b'This is the data, in the clear.'
print('Original    : {} bytes {!r}'.format(
    len(original_data), original_data))

b64_data = base64.b64encode(original_data)
print('b64 Encoded : {} bytes {!r}'.format(
    len(b64_data), b64_data))

b85_data = base64.b85encode(original_data)
print('b85 Encoded : {} bytes {!r}'.format(
    len(b85_data), b85_data))

a85_data = base64.a85encode(original_data)
print('a85 Encoded : {} bytes {!r}'.format(
    len(a85_data), a85_data))

There are several Base85 encodings and different variations are used in Mercurial, git, and the PDF file format. Python includes two implementations, b85encode() implements the version used in Git and Mercurial while a85encode() implements the Ascii85 variant used by PDF files.

$ python3

Original    : 31 bytes b'This is the data, in the clear.'
b64 Encoded : 44 bytes b'VGhpcyBpcyB0aGUgZGF0YSwgaW4gdGhlIGNsZWF
b85 Encoded : 39 bytes b'RA^~)AZc?TbZBKDWMOn+EFfuaAarPDAY*K0VR9}
a85 Encoded : 39 bytes b'<+oue+DGm>FD,5.A79Rg/0JYE+EV:.+Cf5!@<*t

See also