1 GB200 for 12 concurrent users. Even if you get that down to 1 H100 per user, serving billions of people is clearly not shippable at this scale. And that’s before adding voice + conversation stack.
This is why at Animation Inc. we built our on-device animation model, allowing the entire visualization layer to run locally on the users device.
No GPUs cloud required vs. 200,000+ GB200s just for visualisation? 👀
MUSE RESEARCH ALERT:
1/ meet Muse Realtime Avatar, the state-of-the-art model that brings your muse to life alongside Muse Realtime Voice! it animates your muse as it talks to you. voice & video are in sync, so muse will answer in under a second, for as long as you want to chat.