[chefandy]

Seems to me that smaller, more focused models would be more practical. Why would the tulip machine need to know the rules of canasta or the history of peanut butter and jelly sandwiches? I wonder if there's going to be a persistent monolith/microservices cycle.

[senseiV]

well world model in the context of the tulip fields, so models could be finetuned+sheared to drop size and remain effective