Sorry, a noob question here, but if I have 1 Gemini API Key, are requests answered in parallel?
So if I code 5 agents that all send an API request at the same time, do they have to wait sequentially? Or are the requests all processed in parallel as long as I don’t go above RPM?
Welcome to the forum.
The Gemini API is stateless. That means each request is atomic, doesn’t depend on anything that happened before and doesn’t depend on anything happening simultaneously. That guarantees that any number of concurrent requests will all work without you having to worry about sequencing.
Hope that helps.