[hawski]

I'm sure GPT will cover more ground much easier, but aren't LLMs in general much more resource intensive? It seems they will remain expensive to run for quite some time.