LAS VEGAS – Hewlett Packard Enterprise launched a new tool for private cloud AI with Nvidia's top-level GPUs on Tuesday, adding new hardware and agent AI along with expansion of its AI Factory tools.
AI factories are essentially large data centers dedicated to accelerating and developing AI models, from data intake to training and inference. HPE uses its own AI Factory tools along with NVIDIA's Enterprise AI Factory to enable its customers to leverage the platform and build their own AI Factory.
The Houston-based IT infrastructure giant detailed its AI computing portfolio expansion at HPE Discover 2025 this week, promoting NVIDIA's new hardware and software partnership and promoting the accent of financial services agent AI.
“We are committed to providing a range of services to our customers,” said Julia Palmer, Vice President of Research at Gartner. “From a roadmap perspective, HPE is moving beyond progressive improvements.”
Private Cloud AI Gets Boost
For HPE Private Cloud AI, the company has launched a ProLiant Compute Gen12 server with an NVIDIA RTX Pro 6000 Blackwell GPU (now available). HPE uses the new Proliant Compute XD690, which supports Nvidia Blackwell Ultra GPUs, to enhance AI capabilities. The server will be available in October.
Cheri Williams, senior vice president of HPE's Private Cloud and Flex Solutions, said the new launch will help unlock the enterprise possibilities of AI.
“The adoption of generative AI is a little slower than expected a year ago,” Williams said at a press conference. “Data is still scattered everywhere, sovereignty remains uncertain. Customers continue to struggle with how to integrate AI into an already evolving hybrid cloud model.
Public and private clouds that operate seamlessly at hybrid environments and edges can help remove AI guesses and promote focus and faster ROI through measurable and predictable results, she said.
“This is a truly important milestone for us, and more importantly for our customers who rely on seamless, scalable AI operations,” she said.
Steven Dickens, CEO and principal analyst at HyperFrame Research, said HPE's strategy is looking at bigger photos.
“HPE's approach to private cloud AI is holistic and broad,” Dickens said in an email interview. “While Nvidia's hardware components, such as GPUs, are not distinguished by their hardware stack, the overall approach to the curated opinion deployment model is important for clients looking to move towards full-scale production.
“If we combine this curated hardware approach with a robust service portfolio and pre-verified ISV use cases, this is a huge thing for clients looking to reach their ROI immediately.”
In an interview with Informa Tech Target, Williams said that HPE's platform allows businesses to grow AI operations regardless of the speed of hardware advancement.
“We are committed to our enterprise customers that as our roadmap changes, technology evolves and improves, what we invest will be seamlessly integrated with the next generation. So there's that investment protection.”
HPE is trying to meet customers along the way as businesses tackle the dilemma of whether to use on-premises, cloud and hybrid platforms, experts said.
Partnership with Accenture
HPE will also partner with Accenture to create agent AI for financial services and procurement using the Accenture AI Refinery platform built on NVIDIA AI Enterprise and deployed on HPE Private Cloud AI.
“The collaboration with Accenture brings important domain expertise and strategies for a common market, especially in regulated industries,” Gartner's Palmer said. “Partnerships accelerate the deployment of industry-specific agent AI solutions in complex, mission-critical environments.”
Shane Snider is a senior news writer covering Informa TechTarget infrastructure news.
