In an effort to speed up the adoption and implementation of artificial intelligence in businesses, Dell Technologies has unveiled new enterprise AI solutions created in collaboration with NVIDIA.
With NVIDIA, the company introduced improvements to the Dell AI Factory, providing managed services, full-stack enterprise AI solutions, and improved infrastructure for businesses looking to expand their AI operations.
The air-cooled Dell PowerEdge XE9780 and XE9785 servers, which make integration into current enterprise data centers easier, are among the recently announced products.
The liquid-cooled Dell PowerEdge XE9780L and XE9785L versions, which are intended to speed up rack-scale deployment, are a good complement to these. Up to 192 NVIDIA Blackwell Ultra GPUs with direct-to-chip liquid cooling are supported by the new server line, and each Dell IR7000 rack may be customized to hold up to 256 NVIDIA Blackwell Ultra GPUs.
With the 8-way NVIDIA HGX B300, these servers—the next generation following Dell's PowerEdge XE9680—can train huge language models up to four times quicker. The Dell PowerEdge XE9712 with NVIDIA GB300 NVL72 is praised for its five-fold increase in throughput, up to fifty times greater output from AI reasoning inference, and rack-scale training efficiency.
To increase power efficiency on these systems, Dell also introduced PowerCool technology.
The Dell PowerEdge XE7745, which will be offered with NVIDIA RTX Pro 6000 Blackwell Server Edition GPUs in July 2025, is part of the server portfolio expansion. Targeting physical and agentic AI use cases including robotics, digital twins, and multi-modal AI applications, this platform, which is supported within the NVIDIA Enterprise AI Factory approved architecture, can accommodate up to eight GPUs in a 4U chassis.
With a new PowerEdge XE server slated for usage in Dell Integrated Rack Scalable Systems, Dell has also revealed plans to support the NVIDIA Vera CPU and the NVIDIA Vera Rubin platform.
Dell added the PowerSwitch SN5600 and SN2201 Ethernet, which are components of the NVIDIA Spectrum-X Ethernet networking platform, along with the NVIDIA Quantum-X800 InfiniBand switches to its lineup in order to meet the needs of connectivity and networking. These switches support Dell's ProSupport and Deployment Services and have a throughput of up to 800 gigabits per second.
Also Read: How Antonio Neri's Leadership Propelled HPE's Monumental Growth
Improving data management for AI applications is another area of emphasis. The company claims that apps can gain from always-on access to high-quality data thanks to enhancements made to the Dell AI Data Platform. With a denser, software-defined solution, Dell ObjectScale now supports large-scale AI deployments with the goal of lowering costs and data center footprint. Performance and scalability are intended to be increased by integrations with NVIDIA BlueField-3 and Spectrum-4 networking components.