Meta’s Muse agents run in small CPU-only AMD EPYC Turin sandboxes
Meta isolates Muse agents on modest AMD EPYC Turin VMs while relying on separate GPU servers for inference as usage passes 500,000 daily users.
Meta runs its Muse agent in CPU-only sandboxes on AMD EPYC Turin systems, with two cores and 8GB of memory allocated to each sandbox. Evan Hoffman and Tae Kim found that the virtual machines report AMD EPYC 9D25 processors, while Meta uses separate GPU servers for inference. Muse had reportedly exceeded 500,000 daily active users a few days before the report. Hoffman said Muse offered to configure SSH access to its private VM; the article describes a possible route to more harmful commands, but says that has not happened.
Why it matters
Running Muse agents in small, CPU-only sandboxes on AMD EPYC Turin systems means Meta is tightly scoping the compute and isolation model for each user session, even as the service scales past 500,000 daily active users. With separate GPU servers handling inference and optional SSH access offered to a private VM, the design highlights how Meta is trying to balance resource efficiency, isolation and control of what commands agents can execute, while the report notes that more damaging actions had not occurred at the time.
Signal or noise?
Does this story matter, or is it hype? Decide before you see what everyone else thinks.
Sources
- Tom's Hardware