Most llms can equally engage with text in picture form as text in token form. In fact my initial research on this (later corroborated by actual published papers) indicate that this is a cheap way to save on tokens.
That would be my operating assumption at least.