Given typical inference speeds for SOTA models, I rather suspect that you could also do it for inference without latency being particularly notable. We might n

Given typical inference speeds for SOTA models, I rather suspect that you could also do it for inference without latency being particularly notable. We might need to lay a few more cables to New Zealand though.