There is some speculation that GPT-5 uses a router to decide which expert model to deploy (e.g. to mini vs o/thinking models). So the router might decide that t

There is some speculation that GPT-5 uses a router to decide which expert model to deploy (e.g. to mini vs o/thinking models). So the router might decide that the query can be solved by a cheaper model and this model gives worse results.