Phind is fine, except that it's quite unlikely to actually use GPT-4 even when you select "Use Best Model". It takes your query and tries to predict whether it'

Phind is fine, except that it's quite unlikely to actually use GPT-4 even when you select "Use Best Model". It takes your query and tries to predict whether it's highly complex and technical, and will only use GPT-4 if that's the case. It's surprisingly difficult to trigger GPT-4 here. This used to be clearer because Phind would tell you which model it used. Now it says "PHIND MODEL", which muddies the water a bit.