There is really no way to make the ensemble behave with an acceptable level of consistency.
Where we ended up is now having a frontier model generate a whole tree of possible execution plans, and then have the user select one of those path, and then we just run whatever the user chose in a plain sequence until the next decision point that needs user approval.
There is really no way to make the ensemble behave with an acceptable level of consistency.
Where we ended up is now having a frontier model generate a whole tree of possible execution plans, and then have the user select one of those path, and then we just run whatever the user chose in a plain sequence until the next decision point that needs user approval.