Could someone explain in simple terms which of all these models is the most efficient for AG??
The most “Efficient” or like, the one that would consume the less, is the Gemini 3.5 Flash (Low)
The flash models are in theory, the “cheaper ones”.
The "High, “Medium” and “low” refers to the amount of tokens it will use during “Thinking”.
So if you have the “cheap model (Flash)” but you put it on “High” it might use more tokens than the Pro model (more expensive) in the “Low” setting, which is for expending less tokens thinking.
that’s the reasoning behind those names and specifications, but there’s some granularity to it. I’m still testing it so i don’t know at which point one surpasses the other in consumption of token usage/limit rates.