Are you insinuating that your agents are more effective at 20 tok/s than 200 tok/s?

I don't even understand your analogy. I'm not watching the youtube video or reading the book. The model is. And, unlike me, a big computer can watch 20 youtube videos faster than a small computer can watch 1, and do a far better job understanding (bigger model at data center scale is both smarter and faster than smaller model at local scale).