Yes, UTF-8 was up to 6 bytes per character early on. Some broken implementations like MySQL's limit it to up to 3 bytes per character. The actual number is 4.
I think with decomposed Hangul, you can end up with 6 or more bytes per character, due to each part of it being two bytes, and 2-4(?) parts per character.
