I'm trying to get funding to do just this. Essentially nested LLMs were every LLM is very topical and if it's not trained on something there's indexes to other

I'm trying to get funding to do just this. Essentially nested LLMs were every LLM is very topical and if it's not trained on something there's indexes to other LLMs it can ask to get the right answer. I think of it kinda like a hub and spoke model.