A compact, economical single-processor server. It supports a single-width GPU, which is enough for light inference or a small internal service, but it is not an AI training machine and should not be sold as one.
No payment is taken online. Every request is an enquiry answered by an engineer.
Where does this sit?
Team & Departmental. A server holding one to four accelerators, shared by a team or serving a single internal application. Typically bought by an sme or an internal it team. Comms room or a small rack.
Who it's for
What it runs
When it is the wrong answer
Every figure below is explained in plain English. Switch to the technical view for the bare specification.
1U, single-socket rack server
1x Intel Xeon 6
Entry-level single-wide GPU — exact supported model list awaiting verification
DDR5 — maximum capacity awaiting verification
Up to 4-10 SFF or LFF depending on configuration — exact figures awaiting verification
OCP 3.0 plus embedded 1GbE
Air-cooled
Verification
Manufacturer documented — HPE ProLiant DL320 Gen11 QuickSpecs a50004315enw (version 40, July 2026). Reviewed 2026-08-31.
Manufacturer sourceThe things that catch people out after the hardware has already been ordered.
What else will I need?
Relationships documented by the manufacturer, or by us during a deployment.
An eight-accelerator AMD Instinct training server in an 8U air-cooled chassis.
An eight-GPU NVIDIA Blackwell Ultra training server, air-cooled.
A 4U two-socket server taking up to ten double-wide GPUs, sized for inference.