I want to do VLAN Tagging on smaller Boxes like the PowerBox Pro.

With the new Bridge Filtering configuration I do this to get a Layer3 Interface into a vlan:
/interface vlan add interface=bridge1 name=vlan200 vlan-id=200

and then set an IP:
/ip address add address=20.0.0.1/24 interface=vlan200 network=20.0.0.0

So sending Packets into this net/vlan from the PowerBox. Who does the tagging? The CPU of the PoweBox or the included switch HW (QCA8337)?

As the CPU of the PowerBoxPro is not that fast I guess doing the tagging process with CPU would reduce the usable Bandwidth esp with small packets. Looking at the Test Table 64 byte Packet traverse with 267MBit/s max without tagging.