The speaker says a fast model is suitable for the first stage when it only needs to respond to a prompt, because requiring that stage to think can slow the workflow.

“If you're done of thinking then you can actually use any of the like if you want a fast model which responds to like the prompt and does not need to think because if you want it to think you actually start to think about”

“paralyzing the work because you want the first model to be super fast.”