Connects Nebula to local AI model servers on your machine, keeping inference traffic on-device and private.
