Hewlett Packard Enterprise has unveiled a new set of AI factory and supercomputing innovations in collaboration with NVIDIA at NVIDIA GTC 2026, aimed at accelerating large-scale AI adoption across enterprises, research institutions, and sovereign environments.
The announcement focuses on expanding the NVIDIA AI Computing by HPE portfolio, delivering full-stack solutions that integrate compute, GPUs, networking, liquid cooling, software, and services to support high-performance AI workloads at scale.
A key highlight is the advancement of HPE’s supercomputing capabilities with the HPE Cray Supercomputing GX5000, designed to unify AI and high-performance computing (HPC). The platform will support next-generation NVIDIA technologies, including Vera CPU-based compute blades and Quantum-X800 InfiniBand networking, enabling faster scientific discovery and complex workload processing.
According to Trish Damkroger, the integration of AI with traditional HPC workloads is becoming central to accelerating breakthroughs across industries such as life sciences, engineering, and manufacturing.
HPE is also enhancing its AI Factory portfolio with new infrastructure built on NVIDIA’s latest architectures. This includes the NVIDIA Vera Rubin NVL72 rack-scale system, designed for trillion-parameter AI models, and the HPE Compute XD700, a high-density GPU server optimised for both AI training and inference.
These systems are engineered to improve efficiency, scalability, and performance while reducing power and cooling requirements, key considerations for next-generation AI deployments. The solutions are further supported by integrated software capabilities such as NVIDIA Mission Control and expanded multi-tenancy options, enabling organisations to manage large-scale AI workloads with greater flexibility and operational control.
The collaboration also strengthens HPE’s positioning in sovereign AI and neo-cloud environments, where organisations require secure, high-performance infrastructure to manage sensitive data and large-scale AI models.
Chris Marriott emphasised that the partnership brings together accelerated computing, advanced networking, and energy-efficient infrastructure to enable faster time-to-insight for enterprises and governments.
With these advancements, HPE and NVIDIA are reinforcing their joint strategy to deliver production-ready, scalable AI infrastructure, enabling organisations to move from experimentation to real-world deployment across increasingly complex and data-intensive environments.