The Base64 Index Table (64 Characters)

The term "Base64" literally means "base 64". Just like our normal decimal system is base 10 (using the digits 0-9), Base64 uses exactly 64 distinct characters to represent numbers.

When binary data is converted to Base64, the bits are grouped into 6-bit chunks. Since $2^6 = 64$, a 6-bit chunk can hold exactly 64 possible values (0 through 63). Each of these values is mapped to a specific character according to the standard Base64 Index Table.

The Standard Base64 Alphabet

The standard alphabet, defined in RFC 4648, uses the most universally supported and printable characters available in ASCII. It consists of:

• Values 0 to 25: Uppercase letters (A-Z)
• Values 26 to 51: Lowercase letters (a-z)
• Values 52 to 61: Numbers (0-9)
• Value 62: Plus sign (+)
• Value 63: Forward slash (/)

Interactive Base64 Index Table

Use the viewer below to see the exact mapping of decimal values to their respective Base64 characters. You can toggle between Standard Base64 and the Base64URL variant to see how the last two symbols change.

The 65th Character: Padding (=)

Wait, if Base64 only has 64 characters, what is the equals sign (=) that we often see at the end of Base64 strings?

The equals sign is not part of the Base64 index alphabet. It is a special padding character used only at the very end of an encoded string.

Because the Base64 algorithm requires input data to be grouped into 24-bit (3 byte) blocks, if the input data does not perfectly divide by 3 bytes, the algorithm artificially pads the end with zero-bits. It then uses the = character to indicate exactly how many bytes of padding were added, allowing the decoder to correctly trim the output.

• No padding: The data length was a multiple of 3 bytes.
• One =: 2 bytes of input remained.
• Two ==: 1 byte of input remained.