Key takeaways:
-
- Dell Technologies announces next-generation AI infrastructure solutions based on NVIDIA’s Vera Rubin platform, featuring PowerEdge servers with Vera Rubin NVL72 architecture delivering 3.6 exaflops performance.
- New PowerSwitch networking solutions will support Spectrum-6 technology with 102.4 Tb/s switching capacity, while enhanced data processing capabilities include BlueField-4 integration for 800 Gb/s performance and support for NVIDIA Nemotron 3 AI models.
Dell Technologies, the world’s top provider of AI infrastructure¹, supports NVIDIA’s latest platform innovations, delivering unprecedented computing power and efficiency for enterprise AI deployments.
Advancing Dell’s flagship AI server line with NVIDIA Rubin Platform Integration
As noted at Dell Technologies World 2025, Dell is expanding the Dell AI Factory with NVIDIA with the latest NVIDIA Rubin platform – to power the next generation of AI, supercharging agentic AI and AI reasoning. These systems build upon Dell’s market leadership in AI infrastructure, extending the success of the PowerEdge XE9712 with NVIDIA GB300 NVL72, which marked the first market delivery of this groundbreaking technology to CoreWeave.
The upcoming PowerEdge servers will feature the NVIDIA Vera Rubin NVL72 platform, delivering 3.6 exaflops of AI performance. These systems will incorporate 75TB of fast memory and advanced resiliency capabilities within Dell’s proven rack-scale infrastructure.
These updates reflect Dell and NVIDIA’s ongoing collaboration to make enterprise AI more accessible and seamlessly scalable. The new platforms will support the NVIDIA Vera Arm-based CPU, featuring 88 Custom Olympus cores with 176 threads with NVIDIA spatial multithreading capabilities and 1.2 TB/s memory bandwidth, specifically designed as a data movement engine for agentic AI applications.
Expanding HGX Server Capabilities
Expanding our AI server portfolio, Dell PowerEdge will support NVIDIA HGX Rubin NVL8 configurations, delivering an estimated 400 petaflops of AI performance. These platforms will feature 2.3 TB of HBM4 memory, 176 TB/s memory bandwidth and 800 Gb/s NVIDIA ConnectX-9 SuperNICs and NVIDIA BlueField-4 DPUs to power the next generation of AI factories.
The new servers maintain Dell’s focus on practical deployment flexibility to meet diverse data center requirements. This approach ensures organizations can implement high-performance AI infrastructure without extensive facility modifications.
Enhanced Networking and Data Processing
The Dell PowerSwitch portfolio will expand with NVIDIA Spectrum-6 Ethernet networking technology, delivering 102.4 Tb/s switching capacity, and supporting up to 512 ports of 800G connectivity with co-packaged optics (CPO). These CPO switches will provide 5x power efficiency improvements, 10x reliability enhancements, and 5x longer uptime for AI applications compared to switches using traditional pluggable transceivers.
These networking advances complement Dell AI Data Platform innovations, including enhanced capabilities with NVIDIA BlueField-4 data and storage processor integration. BlueField-4 will provide 800 Gb/s networking performance with 2x greater networking and cryptography acceleration than BlueField-3², supporting the NVIDIA Inference Context Memory Storage Platform for KV Cache storage tiering and AI-native data processing capabilities.
Supporting Intelligent AI Models
Dell is also adopting the NVIDIA Nemotron family of open models, spanning from 30B to 500B parameters across Nano, Super, and Ultra configurations. These models’ feature hybrid mixture-of-experts (MoE) architecture and represent a significant advancement in making AI more intelligent and efficient, enabling organizations to freely deploy sophisticated reasoning and inference workloads.
The integration of Nemotron 3 with Dell’s AI Factory with NVIDIA solutions will provide developers with a comprehensive suite of tools for developing and deploying agentic AI agents and applications. Learn more about Dell’s Nemotron 3 solutions.
Meeting Growing AI Infrastructure Demand
The rapid expansion of AI infrastructure requirements across cloud service providers and enterprise environments drives the need for more powerful, efficient computing solutions. Dell’s offerings will address these challenges by providing scalable architecture that can grow with organizational needs while maintaining operational efficiency.
These next-gen innovations position enterprises to capitalize on AI opportunities without the traditional infrastructure limitations that have constrained deployment scale and performance. The combination of advanced computing, networking, and data processing capabilities creates a foundation for the next era of the Dell AI Factory with NVIDIA.
Dell Technologies continues to lead enterprise AI infrastructure innovation, delivering solutions that transform AI potential into measurable business outcomes. Our partnership with NVIDIA ensures customers receive cutting-edge technology with the reliability and support needed for critical business applications.
To learn more, visit the Dell AI Factory with NVIDIA page.
1IDC Worldwide Quarterly Artificial Intelligence Infrastructure Tracker, 2025Q2
2Projected performance gains from previous generation products based on NVIDIA’s own design estimates and subject to change.
