Hacker News Reader

Congrats on the launch! Looks fantastic.

I've been thinking about exactly this kind of architecture (vision-language models + physical robot ==> performing tasks)

I'd love to tinker with one (or more) of these 'Bots.

Question: Is the entire inference in the cloud? or on the Bot's hardware?

View on HN

It's split between the bot and the cloud! High-level decision making and training is performed in the cloud for performance reasons, and low-level control and inference of VLAs is performed on the robot which has a GPU onboard!

And yeah it's a powerful architecture, but it's not enough if you want to perform task, the VLMs orchestrate but you need another model for manipulation. And we put all of these together :)

Happy to chat further!

Reply on news.ycombinator.com