INFRA Signal 142
HP ZGX Fury now orderable with GB300 Superchip, Red Hat AI Factory platform in development
↗
HP's ZGX Fury AI workstation, built on NVIDIA's GB300 Grace Blackwell Ultra Desktop Superchip with 748GB of unified memory, is now available to order, and HP is collaborating with Red Hat and NVIDIA to bring Red Hat AI Factory to the platform as managed edge inference infrastructure.
The ZGX Fury puts trillion-parameter inference and fine-tuning of 100 billion parameter class models on a single box without a data center, and the Red Hat AI Factory integration adds workload isolation, governance, and multi-workload orchestration that IT teams need to manage it as edge infrastructure rather than a personal workstation.
Written by elseif from the cluster below · every claim links back to a sourceThe three things worth knowing
The GB300 Grace Blackwell Ultra Desktop Superchip combines a 252GB HBM3e GPU with a 72-core Grace CPU's 496GB LPDDR5X over NVLink-C2C into a 748GB coherent addressable space.
HP is positioning the ZGX Fury as shared edge inference infrastructure for departments, factory floors, and branch offices rather than as a personal workstation.
HP and Red Hat are developing an enterprise-grade AI platform running Red Hat AI Factory with NVIDIA on the ZGX Fury, with RHEL certification already listed in the Red Hat Ecosystem Catalog.
THE CLUSTER