HPE Launches AI Factory Solutions Powered by Blackwell GPUs and Gen12 Servers

HPE simplifies AI adoption with integrated, end-to-end solutions and services.

At the HPE Discover 2025, Hewlett Packard Enterprise (HPE) unveiled a comprehensive suite of solutions designed to accelerate the creation, adoption, and management of AI factories across the entire AI lifecycle. These solutions are tailored to meet the needs of virtually all organizations, regardless of size or industry.

HPE is expanding its NVIDIA AI Computing by HPE portfolio with the addition of NVIDIA Blackwell GPUs. This expansion introduces new, modular solutions optimized for service providers, model builders, and sovereign entities. It also includes the next-generation HPE Private Cloud AI, a turnkey AI factory solution designed for enterprises.

By offering integrated, end-to-end solutions and services, HPE simplifies the customer experience, eliminating the complexity of creating a modern AI-ready data center from various components. This approach enables organizations to focus on innovation rather than managing the integration of infrastructure.

Infrastructure as the Foundation for AI

Antonio Neri, President and CEO of HPE, emphasized that the transformative potential of generative, agentic, and physical AI hinges on robust infrastructure and high-quality data. He noted that organizations must possess the correct data, intelligence, and strategic vision to fully capitalize on the opportunities presented by AI. Neri highlighted the partnership between HPE and NVIDIA as a cornerstone in delivering comprehensive AI infrastructure, empowering businesses to achieve sustainable value and drive meaningful societal change.

Jensen Huang, Founder and CEO of NVIDIA, echoed this sentiment, describing the emergence of a new industrial era characterized by the generation of large-scale intelligence. He underscored the collaboration between HPE and NVIDIA as instrumental in providing the infrastructure necessary for companies to harness their data, accelerate innovation, and maintain a competitive edge in the rapidly evolving AI landscape.

HPE Private Cloud AI Speeds Time-to-Value

At the heart of the NVIDIA AI Computing by HPE portfolio is HPE Private Cloud AI, a fully integrated, turnkey AI factory. This solution incorporates NVIDIA accelerated computing, networking, and software, delivering a seamless experience for enterprise AI deployment. Private Cloud AI supports NVIDIA Blackwell accelerated computing through HPE ProLiant Compute Gen12 servers, which are recognized for their leadership in over 23 AI benchmarks. These servers provide secure boundaries to prevent tampering, utilize post-quantum cryptography, and offer trusted supply chain capabilities at both the rack and server levels.

The platform ensures investment protection and smooth scalability across GPU generations, featuring support for NVIDIA H200 NVL and the latest NVIDIA RTX PRO 6000 Server Edition GPUs. This flexibility accommodates a wide range of enterprise AI workloads, including agentic and physical AI applications. A new federated architecture enables efficient resource pooling, allowing all AI workloads to leverage the latest GPUs and resources.

HPE ProLiant Compute Gen12 with AMD EPYC 9005 Processors

HPE unveiled a new ProLiant Compute Gen12 portfolio with two AMD-powered servers. The HPE ProLiant DL325 Gen12 and DL345 Gen12, featuring 5th Gen AMD EPYC processors, are designed to deliver next-level security, performance, and energy efficiency, meeting the evolving needs of hybrid cloud and data-driven workloads. Building on the momentum of HPE’s earlier Intel Xeon 6-based ProLiant Compute Gen12 launch, these AMD EPYC models provide customers with additional flexibility to power modern applications, such as AI, edge computing, virtualization, and VDI, while driving greater operational efficiency across data center and edge environments.

The HPE ProLiant Compute Gen12, featuring 5th Gen AMD EPYC processors, is designed to address many of the top challenges facing IT organizations today. From improving server security to boosting efficiency and supporting emerging workloads, these AMD-based servers offer a well-balanced platform of performance, scalability, and flexibility for enterprise environments.

The DL325 Gen12 and DL345 Gen12 feature multilayered security, including HPE’s silicon root of trust and AMD’s Infinity Guard, providing organizations with stronger protection from silicon through software. The Gen12 platform is also designed with compliance readiness for emerging quantum-resilient encryption standards.

Performance gains are notable, with up to 41% better performance per watt compared to Gen10 servers. These improvements help organizations run modern AI, virtualization, and edge workloads with lower power consumption and reduced operating costs. Support for higher-wattage CPUs (up to 500W) and faster DDR5 memory further enhances performance for demanding applications.

Management is strengthened through HPE Compute Ops Management, providing greater visibility and control across global server fleets. AI-driven insights enable IT teams to optimize performance, predict energy usage, and streamline lifecycle management at scale.

DL325 Gen12

The HPE ProLiant Compute DL325 Gen12 is a 1U, single-socket server designed for hybrid cloud, containerized workloads, and secure edge deployments. It delivers high compute density and efficiency, powered by 5th Gen AMD EPYC processors.

Key enhancements include:

Single AMD EPYC 9005 Series processor support with up to 192 cores total
Up to 6TB DDR5 memory across (2 DIMM per channel) for maximum performance
Flexible storage with support up to 300TB capacity
GPU support with up to 2 double-wide GPUs for accelerated workloads
PCIe Gen5 expansion with up to two slots and dual OCP 3.0 slots
Available direct liquid cooling (DLC) for high-performance thermal management
Integrated iLO 7 for management with enhanced UX

DL345 Gen12

The HPE ProLiant Compute DL345 Gen12 is a 2U, single-socket server optimized for compute and data-intensive workloads, including AI, ML, telco, SDS, and database applications. It combines scalable performance with flexible storage and GPU capabilities.

Key enhancements include:

Single AMD EPYC 9005 Series processor support with up to 192 cores total
Up to 6TB DDR5 memory across (2 DIMM per channel) for enhanced bandwidth
Extensive storage support up to 552TB capacity
Versatile GPU support with up to 2 double-wide 400W GPUs or 6 single-wide GPUs
PCIe Gen5 expansion with up to six slots and dual OCP 3.0 slots
Available direct liquid cooling (DLC) for high-performance thermal management
Integrated iLO 7 for management with enhanced UX

These new AMD-based ProLiant Gen12 servers give HPE customers a broader set of choices to modernize their infrastructure and meet the growing demands of today’s hybrid and AI-powered environments.

For organizations with stringent data privacy requirements, Private Cloud AI offers air-gapped management. Multi-tenancy capabilities enable enterprises to partition and allocate resources across teams, encouraging collaboration while maintaining security. The latest NVIDIA AI Blueprints, including the NVIDIA AI-Q Blueprint for AI agent creation and workflow automation, are integrated into the offering. Additionally, a “try-and-buy” program allows customers to evaluate Private Cloud AI within the Equinix global network before making a purchase commitment.

AI Solutions for Financial Services

HPE is collaborating with Accenture to develop AI solutions tailored for the financial services sector. Together, they are launching a joint go-to-market offering that leverages the Accenture AI Refinery platform, built on NVIDIA AI Enterprise and deployed on HPE Private Cloud AI. HPE is already implementing this solution within its finance organization, exploring applications in spend management, strategic relationship analysis, and contract obligation management. This partnership aims to accelerate AI adoption and drive digital innovation in one of the world’s most regulated industries.

HPE Financial Services (HPEFS) has introduced a new program that enables enterprises to begin their AI projects with reduced payments for Private Cloud AI during the first six months. HPEFS also provides a comprehensive suite of customizable financing and IT lifecycle services, including programs that allow customers to leverage their existing technology assets as capital for additional AI initiatives. This approach accelerates the transition to AI, enabling organizations to innovate without financial barriers.

Comprehensive AI Factory Portfolio and Validated Solutions

HPE’s AI factory portfolio now includes new validated solutions that integrate the latest AI innovations and expertise. Drawing on five decades of leadership in liquid cooling, HPE offers HPE Morpheus Enterprise Software for unified control and HPE OpsRamp Software for comprehensive observability. Notably, HPE OpsRamp is a validated observability solution for the NVIDIA Enterprise AI Factory, ensuring robust monitoring and management capabilities.

These composable solutions are provided as complete technology stacks that are pre-integrated to accelerate the time-to-value for customers. The AI factory, scaled for large organizations such as service providers and model builders, includes HPE ProLiant Compute XD, NVIDIA AI Enterprise software and blueprints, advanced cooling technologies, and a comprehensive range of advisory and professional services. For sovereign entities, including governments and public sector organizations, HPE offers specialized solutions with air-gapped management and features that support data, technological, and operational sovereignty.

By leveraging NVIDIA’s Enterprise AI factory-validated design, HPE’s solutions utilize the latest NVIDIA accelerated computing, Spectrum-X Ethernet networking, BlueField-3 DPUs, and AI Enterprise software. This combination delivers high performance, robust security, efficient storage acceleration, and a scalable infrastructure for the next era of AI.

HPE Compute XD690: Powering the Next Generation of AI

New HPE Compute XD690 supports eight NVIDIA Blackwell Ultra GPUs and reflects HPE’s continued commitment to be a time-to-market leader with NVIDIA’s latest innovations in accelerated computing for AI. HPE Performance Cluster Manager delivers fully integrated systems management and advanced infrastructure monitoring and alerting across large, complex AI environments that scale thousands of nodes.

Expanding the Unleash AI Ecosystem

HPE’s Unleash AI partner ecosystem has grown to include 26 new partners, enabling support for more than 75 AI use cases. Through this expanded ecosystem, the latest NVIDIA AI Blueprints, and HPE’s internal professional services, customers gain access to solutions spanning agentic AI, sovereign AI, smart cities, industrial and manufacturing applications, data governance and privacy, responsible AI, video analytics, security, and cybersecurity.

HPE Alletra Storage MP X10000: Fueling AI Factories with Data

To ensure that AI factories and use cases are continuously supplied with AI-ready unstructured data, HPE Alletra Storage MP X10000 will support Model Context Protocol (MCP) servers. Integrating MCP with the X10000’s built-in data intelligence accelerates data pipelines and allows AI factories, applications, and agents to process and act on intelligent unstructured data.

Accelerating and Simplifying the AI Journey

HPE alleviates the burden of building and adopting AI factories, accelerating the time to value through new services that support customers on their AI journey. These services encompass designing, financing, building, and securing AI factories from Day -1 to Day 2+, focusing on efficient and sustainable operations. The services also include model development and migration, business benefit analysis, deployment, education, management, ongoing support, and tech refreshes.

Engage with StorageReview