Hello,
currently you can regenerate an existing query with the same model and then move between these generations using the <1/2> buttons. Or you can ask something else with a different model using the same context. It would be very useful to be able to regenerate a reply not with the same model but be able to choose a completely different model. So you could have the same prompt but generate reply 1 with model 1, reply 2 with model 2 and so on...
It would expand usefullness for a thread without losing context. Other providers like perplexity also allow this.
