I have had this question. How much better would common LLMs (Llama, GPTN) be if they were only trained in one language? I have to assume they would perform better, but I might be wrong.
they would perform worse, i promise you
This is Hackernews, I would have expected data, not promises.