Dell and AMD Are Expanding What’s Possible for On-Premises AI

Key takeaways: Dell PowerEdge XE7745 and R7725 will support AMD Instinct MI350P PCIe GPUs starting in July 2026, and have expanded the Dell AI Platform with AMD, enabling enterprises to run generative and agentic AI within their existing data center infrastructure — no redesign required.

AI for enterprise is no longer an option, it’s an imperative. The right on-premises solutions offer a competitive advantage providing better performance, control and security. Dell and AMD are delivering enterprises a better option.

Dell PowerEdge servers will support AMD Instinct MI350P PCIe GPUs, equipping enterprises with a performant, cost-effective option to scale agentic and generative AI deployments. In addition, Dell is enhancing the Dell AI Platform with AMD to help scale AI workloads from pilot to production.

What This Means for Enterprise AI

Dell PowerEdge servers with AMD Instinct MI350 PCIe GPUs deliver:

• • Drop-in deployment: Standard air-cooled Dell PowerEdge XE7745 and R7725, no data center redesign required.
  • Leading performance: Up to 4,600 peak teraflops (MXFP4) and 144GB HBM3e memory¹—the highest capacity currently available in a PCIe card accelerator.
  • Broad workload support: Optimized for small, medium and large model inference, RAG pipelines and agentic AI.
  • Open software stack: AMD enterprise AI software integrates with PyTorch, TensorFlow, vLLM and other frameworks — no licensing fees, minimal code rewrites.

A Modular Platform Built to Scale

The Dell AI Platform with AMD’s new modular architecture is designed to grow with enterprise AI needs. It’s designed to provide flexibility, scalability and modularity. Organizations can start with configurations that meet current workloads and expand compute and GPU density over time — without rearchitecting. The platform uses AMD Enterprise AI Suite, AMD ROCm and AMD Inference Server to support the full AI lifecycle: training, fine-tuning, inferencing and agentic workflows, all within a validated, secure, on-premises environment.

For the most demanding workloads, Dell also supports PowerEdge XE9785 servers with AMD MI355X GPUs and EPYC CPUs — purpose-built for foundation model development and large-scale inference.

To learn more about Dell AI Platform with AMD and the new modular architecture, visit www.dell.com/en-us/lp/amd-ai.

¹AMD Scaled data formats conform to OCP standard MX data types.