Ah, you're probably one to ask. They say "queries on some topics will instead receive a response from our next-most-capable model, Claude Opus 4.8." Are they transparent about when that happens, and is it priced at the rate of the underlying model?

They are transparent about when it happens but no reason why. To be fair, it doesn't interrupt the flow, just drops to Opus and proceeds. The most frustrating thing is that it happened on a plan and Fable just refused to have anything to do with the plan.