A secure foundation to power new workloads while streamlining operations and cutting energy use. Gen12 spans Intel® Xeon® 6 rack and tower systems, NVIDIA accelerated platforms, and the NVIDIA® Vera CPU–based DL394 for agentic AI — all wrapped in the security, manageability, and lifecycle tooling enterprises trust from HPE ProLiant.
HPE ProLiant Compute Gen12 meets federal security certification standards, boosts IT productivity with AI-driven insights, and drives major power savings — right-sized for AI, data analytics, edge, hybrid cloud, and VDI.
HPE iLO 7 adds a dedicated secure-enclave security processor — the first server with quantum computing-resistant readiness and FIPS 140-3 Level 3 cryptographic certification.
Up to 41% better performance per watt and up to 65% power savings per year, with optional direct liquid cooling on Intel-based one- and two-socket rack servers.
HPE Compute Ops Management secures and automates server environments with AI-informed insights, map-based visibility, and third-party tool integration.
Rack-dense compute from 1U single-socket to 4U accelerated platforms, plus the Vera CPU–based DL394 built for agentic AI. Pick a model below or have us spec it for you.
2U server powered by the NVIDIA® Vera CPU — built for agentic AI and reinforcement learning, with 88 Olympus cores and up to 1.2 TB/s memory bandwidth. Explore the DL394
1U single-socket rackmount that maximizes rack utilization while mitigating virtualization risks in power-constrained environments. View models
2U single-socket server with reduced power draw and better thermals for data- and storage-centric workloads that need more PCIe lanes. View models
Compact 1U 2P server delivering exceptional compute performance, memory density, scalability, and high-speed data transfer for demanding applications. View models
Two-socket scalable server that powers data analytics and supports hybrid cloud workloads. View models
4U accelerated platform with 6th Gen Intel® Xeon® (64–144 core), up to 4 TB RDIMM, SFF NVMe and EDSFF storage, and up to 8 M-CRPS power supplies. Contact us to configure
NVIDIA Grace CPU and Hopper GPU platform with up to 8 EDSFF NVMe Gen5 drives and up to 4 HPE 1800W–2200W Flex Slot Titanium power supplies. Contact us to configure
Tower form factors for offices and edge sites that need datacenter-class capability without a datacenter.
Every Gen12 server inherits the same foundation, so the choice between models is about form factor and workload — not about giving up protection or manageability.
Processor family, memory, storage, networking, GreenLake consumption, and lead times differ across the DL and ML lineup. Tell us what your workloads need to do, and we'll land you on the right Gen12 build, get you pricing that makes sense, and stay on the order until it's racked.
Six of the eight new Intel Xeon 6–based Gen12 servers ship first: DL320, DL340, DL360, DL380, DL380a, and ML350 Gen12.
HPE Synergy 480 and HPE ProLiant Compute DL580 Gen12 servers are expected to follow.
The full Gen12 portfolio is available as a standalone purchase or as a service via HPE GreenLake.
Tell us about your workloads and a ServerComputeWorks specialist will help you choose, configure, price, and order the right HPE ProLiant Gen12 system.
Talk to an Expert*Intel and Xeon are trademarks of Intel Corporation. NVIDIA, Vera, Grace, and Hopper are trademarks and/or registered trademarks of NVIDIA Corporation. SPEC and SPECrate are registered trademarks of the Standard Performance Evaluation Corporation. FIPS 140-3 Level 3 certification is a standard adopted by NIST and CNSA 2.0. Performance and power-savings claims are based on internal HPE measurements and SPECrate2017_int_base results published on spec.org. All other third-party trademark(s) is/are property of their respective owner(s). Specifications and availability are subject to change.