As far as I know, non-ASCII newlines and whitespace are not supported in Python source code, so there is not a big need to support it in bytes.fromhex() either. But since bytes.fromhex() accepts Unicode strings, I think non-ASCII whitespace would be okay if it was easy to implement.

Serhiy: Whitespace is not treated specially by the base-64 decoders, it is just treated like any non-alphabet character. See <https://docs.python.org/3/library/base64.html#base64.b64decode> and b64decode(validate=True).

Regarding hex-codec, I doubt many people use it in Python 3. To decode a whole string, binascii.unhexlify() or bytes.fromhex() is probably more obvious, and I think the incremental decoder never worked properly (Issue 20132).