Tayschrenn
Young grasshopper
It's the network latency that kills it a bit. Even on a dedicated VLAN with dedicated 1gb network ports (my BI 'server' has a 4-port Intel NIC) it's not as fast as running locally =(By running the AI server on a dedicated machine, you can improve performance for applications like object detection, facial recognition, or image processing without burdening your main system.