HPE unveils four ProLiant Gen13 servers built on 6th Gen AMD EPYC CPUs

Four servers for different workloads

The first wave of ProLiant Gen13 consists of the HPE ProLiant DL585a, DL525, XD245 and XD285 servers. According to HPE, combining its own engineering with 6th Gen AMD EPYC processors delivers higher core density, memory bandwidth and I/O capacity.

The HPE ProLiant DL585a Gen13 is a high-density platform designed to support up to eight double-wide PCIe GPUs, two 6th Gen AMD EPYC processors and PCIe Gen6 connectivity. HPE positions it for demanding workloads such as AI inference, virtualization, retrieval-augmented generation (RAG) model fine-tuning, analytics and emerging agentic AI.

The HPE ProLiant DL525 Gen13 is a 1U, single-socket system with one 6th Gen AMD EPYC processor offering up to 256 cores. It has 16-channel memory, supports DDR5 and MRDIMM, and offers PCIe Gen6 connectivity. HPE names fraud detection, market-data processing, risk and pricing models, electronic design automation (EDA), simulation and AI inference as target uses.

For large-scale, compute-intensive applications, HPE added the ProLiant XD245 and XD285. Both are OCP-compliant and combine high-density compute with shared power and cooling. The XD245 uses a one-Open Rack Unit, four-node direct liquid-cooled design and is powered by up to eight 6th Gen AMD EPYC processors. The XD285 uses a two-Open Rack Unit, two-node air-cooled design, which HPE says extends the architecture to a broader range of data center environments. HPE lists computer-aided engineering, weather and climate modeling, scientific research and financial modeling among their uses.

HPE said Gen13 will expand beyond these systems, with further options for mainstream enterprise, edge, virtualization, software-defined infrastructure, database and AI workloads.

Security and management features

HPE frames the launch around trust. It cites McKinsey's 2026 AI Trust Maturity Survey, which found that nearly two-thirds of respondents named security and risk concerns as the top barrier to fully scaling agentic AI.

ProLiant Gen13 pairs automatic security from HPE Integrated Lights-Out 8 (iLO 8) with expanded HPE Compute Ops Management. HPE describes several features as industry firsts:

• Automatic self-encrypting drive (Auto-SED) capabilities, which protect data at rest from first boot without added configuration, external key managers or manual activation.
• Multi-person authorization, which requires two authorized users to approve high-impact actions such as key provisioning and secure erase.
• Expanded post-quantum cryptography readiness that now extends beyond firmware to help protect server communication paths.

HPE also lists enterprise-grade component verification, built on its Silicon Root of Trust, which validates the authenticity and integrity of critical components as they are added, replaced or modified.

On the management side, Compute Ops Management adds low-touch onboarding and near real-time power capacity insights. HPE says organizations using it spent 50% less time managing servers and generated 209% ROI.

Reliability claims, availability and financing

HPE says each new ProLiant generation undergoes more than 1,000 tests and 10,000 hours of validation. It also states that reliability has been the leading ProLiant customer purchase driver for four consecutive years, with 81% citing reliable performance as a key differentiator from other server vendors. HPE Tech Care Service, the company adds, delivers up to 97% less unplanned downtime. These figures are HPE's own.

Availability is staggered. The DL525 will be available later in 2026, the DL585a in early 2027, and the XD245 and XD285 in 2027. The servers can be financed through HPE Financial Services' 90/9 Advantage program, which offers no payments for 90 days, followed by nine months at 1% per month.

What it means for buyers

For enterprises planning AI and inference deployments, the launch gives HPE's AMD-based server line a new generation with PCIe Gen6, high core counts and built-in security controls that require little manual setup. Krista Satterthwaite, HPE's senior vice president and general manager for Compute, said customers need to turn agentic AI and inference investments into business value "without adding operational complexity or risk."

Since the first system is not due until later in 2026 and the rest follow in 2027, the announcement is mainly a roadmap for organizations planning upcoming infrastructure purchases.