RSA algorithm you can print on a shirt - and that would give you encryption that works as well as it possibly can. There's no "RSA 6.7 Turbo Pro EncryptingPlus" that would give you something radically new that the "t-shirt RSA" can't.

LLMs are massive, complex, and a moving target. Making them is hard, and there's no "we're done" there - the frontier updates every few months. It's easy to get and run a 2B model locally, but what concerns people isn't a cheap end device 2B - it's the 10T RLVR-fried monsters that frontier labs play with nowadays.