it looks like all AI is following the same pattern as GPT-3 again, building bigger models to achieve better results.
Which is actually good news if it still works
Which is actually good news if it still works