Lucas Beyer says Thinky-style robotics would be an interesting direction and asks whether inference API latency after fine-tuning could hurt users far from the servers. He adds that more servers could fix it. Physical Intelligence announced the beta release of its Pi API, opening access to more teams adapting its models for robots and applications.
Thinky but for robotics? I think that's a pretty interesting direction!
I wonder if the latency of the inference API after the fine-tuning might hurt if you're not close to their servers? Nothing that can't be fixed with more servers of course, but I'm curious.
Announcing the beta release of the Pi API. We’ve partnered with companies across diverse industries to adapt our models to their robots and applications, from harvesting strawberries to assembling servers. Now we’re opening access to more teams.
Source: Lucas Beyer (bl16) · x.comPublished · added here
