Dell Technologies said on Monday it has unveiled a new server product, the Dell PowerEdge XE8812, which adopts Nvidia Vera Rubin NVL4 architecture.

The company said the Dell PowerEdge XE8812 can support up to 144 GPUs per rack and will be added to the Nvidia-based Dell AI Factory product lineup.

It is suited to HPC and AI workloads.

Dell plans to use the new product to accelerate expansion of Dell AI Factory across a range of fields worldwide, including advancing sovereign AI initiatives as well as engineering and design workflows and genomics.

The company said that as AI and HPC simulation workloads converge, incremental infrastructure upgrades alone are no longer sufficient to handle the scale and speed of such work.

Arun Narayanan (아룬 나라야난), senior vice president of compute and networking at Dell Technologies, said, "Institutions conducting humanity's most important research, from deciphering the human genome to modeling future energy systems and building sovereign AI infrastructure for national operations, must have infrastructure befitting their projects. The Dell PowerEdge XE8812 reflects Dell's commitment to pushing the boundaries of what is possible, providing computing density, memory and open architecture that can take on workloads once considered impossible."

The Dell PowerEdge XE8812 is scheduled for an official global launch in early 2027.