I posted this video, i. did not spend too much time getting into the details of tuning other than pod resource allocation, i setup an orchestrator an end point and 6 pods of a basic model and the response time bottle neck is what I’m trying to figure out, any help would be appreciated.
