MicroEMACS supports reading and writing text files encoded with UTF-8, a byte-oriented encoding of Unicode. UTF-8 can be thought of as a superset of ASCII: bytes less then 0x80 are identical to ASCII, and bytes greater than or equal to 0x80 are always parts of UTF-8 sequences. UTF-8 characters vary in length from one byte (ASCII) to six bytes, though four bytes is the longest typically seen.

Internally, Microemacs stores lines of text in their original UTF-8 encoding, but displays multibyte UTF-8 sequences to the user as single Unicode characters in the range 0x80 to 0xffff. Unicode characters greater then 0xffff are not supported; in practice these characters are extremely rare.

For the most part, MicroEMACS is able to display UTF-8 characters correctly, with the exception of non-spacing or combining characters. The display of these characters is undefined: sometimes they show up as modifiers of subsequent characters, or as blanks, or as a lowercase 'x' with a modifier.

On Linux, most terminal programs support a standard method for entering Unicode characters at the keyboard: hold down Ctrl and Shift, then press and release 'u', then release Ctrl and Shift, then enter the hex digits of the Unicode character followed by the Enter key. This method works when entering characters at prompts in the echo line, or while entering text in the edit buffer.

If this method doesn't work, you can still enter Unicode characters in the edit buffer using the following command:

M-C-U (unicode)

This command prompts you to enter a line of text containing the hexadecimal values of one or more Unicode characters. The hex values must not have a '0x' prefix, or any other prefix, and must be separated by spaces. MicroEMACS will then insert the corresponding UTF-8 characters into the current buffer.

As an example, entering the string e0 e1 e2 would insert the characters àáâ into the buffer.