Yes, mainly because of the work of experts which went into GPT4.

For example, Llama is nowhere close (even if it's pretty good).

You can think of GPT4 as a way to flexibly access a lot of knowledge from domain experts. Sure, sometimes that flexibility hallucinates things, but it mostly works and we can verify a large part of it.