Comment by shaewest

It's running a very small, non-reasoning model at the moment. But more generally, almost all LLMs argue on the hardware/model they are/are on.