HPE announced updates to HPE AI Factory with NVIDIA today at NVIDIA GTC Berlin, and most of the changes land in HPE Private Cloud AI. Developers, independent software vendors, and AI agents get governed access to the platform’s inference, models, and pipelines through production APIs and Model Context Protocol (MCP) servers, clusters can now mix NVIDIA H200 and RTX PRO 6000 Blackwell Server Edition GPUs, and HPE is introducing the NVIDIA Vera CPU Rack for the agentic workloads that run on CPUs.
HPE AI Factory with NVIDIA bundles models, infrastructure, data, and workflows so enterprises can run AI under centralized security, governance, and operational control. The GTC Berlin updates follow the agentic AI additions HPE made at Discover in June, and they target the same customers: organizations in hybrid, regulated, and sovereign environments that want autonomous AI workloads in production without giving up control of the data and the infrastructure.
“AI agents are becoming an essential part of how enterprises operate and create value,” said Cheri Williams, SVP and GM, Private Cloud and Flex Solutions, HPE. “As organizations deploy AI across industries and increasingly sensitive environments, they need infrastructure that gives them greater control over security, sovereignty, and how enterprise data is accessed and used.”
HPE Private Cloud AI Adds Governed APIs and MCP Servers
The new HPE AI Essentials capabilities give developers, ISVs, and AI agents access to Private Cloud AI’s inference, models, knowledge services, and AI pipelines through production-grade APIs and MCP servers, all under centralized identity management and governance. Developers and ISVs can build Private Cloud AI services into applications and industry-specific solutions, and the MCP servers extend the same access to autonomous agents, with versioning and governance scoped by role, tenant, and project. Audit logging records what agents do inside enterprise systems.
OpenAPI is the source of truth for every supported interface, so documentation, software development kits, and code samples are generated from the same definitions. Role-based access control, machine identity support, and audit logging apply to human and agent access alike. HPE also outlined API-first and Kubernetes-assisted onboarding for connected and air-gapped Private Cloud AI deployments, aimed at organizations with regulated or sovereign deployment requirements.
Mixed-GPU Clusters and Workload Scheduling
HPE Private Cloud AI now runs NVIDIA H200 and RTX PRO 6000 Blackwell Server Edition GPUs in the same cluster, so a customer can match each workload to the GPU that fits it under one operating and governance environment. That capability is generally available, as is NVIDIA’s open-source KAI Scheduler, which adds quota-based and priority-based scheduling across projects to improve GPU utilization and workload placement. Our RTX PRO 4500 Blackwell Server Edition review covers the single-slot member of the same GPU family.
Schneider Electric plans to combine HPE Private Cloud AI with its industrial and operational technology expertise to support what it calls Industrial AI Agents at the Edge, putting inference closer to where operational decisions are made while keeping control over sensitive data. HPE and Schneider Electric have not said which sites or workloads come first.
NVIDIA Vera CPU Rack Targets Agentic Workloads
HPE is introducing the NVIDIA Vera CPU Rack by HPE for the CPU-heavy side of agentic AI and reinforcement learning: agent sandboxes, RL evaluations, and data-processing pipelines that run beside the GPU infrastructure. The liquid-cooled rack carries 256 Vera CPUs and supports more than 22,500 sandboxes. HPE claims 80 percent faster agentic CPU performance without naming a baseline; NVIDIA’s Vera CPU Rack page frames the same figure against traditional CPU infrastructure, also without a test configuration.
NVIDIA’s render of the Vera CPU Rack with a compute tray pulled out
Availability
The Private Cloud AI updates, including administrator-enabled developer and ISV access through the HPE AI Essentials API platform, are scheduled to roll out in late 2026. A self-service developer experience covering application and agent registration, API and MCP discovery, and credential management is planned for 2027. Mixed H200 and RTX PRO 6000 Blackwell Server Edition clusters and KAI scheduling are generally available now.
Amazon