Wouldn't this introduce new economics into the LLM market?
I.e. if the "thinking loop" budget is parameterized, users might pay more (much more) to spend more compute on a particular question/prompt.
I.e. if the "thinking loop" budget is parameterized, users might pay more (much more) to spend more compute on a particular question/prompt.
Given the need for chain-of-thoughts, and that would be budgeted as output, the new model will not be cheap nor fast.
EDIT: Pricing is out and it is definitely not teneable unless you really really have a use case for it.