HP Inc. (NYSE: HPQ) announced a collaboration with Red Hat and Nvidia to develop an enterprise AI platform designed to run AI inference workloads at the edge, closer to users, applications, and data sources.

The planned solution centers on the HP ZGX Fury, a workstation powered by the Nvidia GB300 Grace Blackwell Ultra Desktop Superchip, combined with Red Hat AI Factory with Nvidia. HP states the platform is designed to deliver up to 20 PFLOPS FP4 AI performance for local inference workloads.

HP ZGX Fury is currently available to order and is certified to run on Red Hat Enterprise Linux, with listings available via the Red Hat Ecosystem Catalog. HP said customers will also be able to evaluate a planned fuller solution in a sandboxed environment on HP devices running Red Hat AI Factory with Nvidia, though timing and eligibility details have not yet been disclosed.

According to HP, the platform is intended to help organizations address latency, data privacy, sovereignty, and connectivity concerns by running AI workloads locally rather than relying solely on cloud infrastructure. The system is being designed to support multiple AI workloads simultaneously while maintaining workload isolation and governance controls.

"Together with Red Hat and Nvidia, HP is extending enterprise AI from the data center to the edge with an open, enterprise-grade inference platform designed to give customers greater choice, control and consistency as they deploy local AI factories," said Jim Nottingham, Senior Vice President and Division President, Advanced Compute and Solutions, HP Inc.

HP identified target use cases across manufacturing, healthcare, government, retail, and software development environments, including computer-vision inference near production lines and local AI tools for developers.

The platform is built on Red Hat Enterprise Linux and Red Hat OpenShift and incorporates Nvidia AI Enterprise software and Nvidia CUDA libraries for GPU scheduling and multi-GPU workload orchestration.