AMD Instinct MI350P PCIe GPUs: Run Enterprise AI on Your Existing Infrastructure

As organizations adopt AI, many discover that their infrastructure struggles to keep up. Running AI in the cloud is an option, but the cloud can introduce privacy concerns and unpredictable costs. Upgrading on-prem infrastructure is another option, but supporting large GPU-accelerator platforms can require expensive redesigns in data center power and cooling.

Our new AMD Instinct™ MI350 PCIe®  cards give your enterprise a third option: Leadership AI performance designed to fit the data center infrastructure you already own.

Performance That Drops into Your Existing Racks

Designed to help you prepare for the agentic AI era, AMD Instinct MI350P PCIe cards are dual-slot drop-in cards for standard air-cooled servers. They are built to deploy inference on premises within your current data center’s power, cooling and rack infrastructure. AMD Instinct GPUs in cost-effective PCIe cards round out the AMD AI compute portfolio, providing a range of options for your enterprise as it navigates its unique AI adoption curve.

The PCIe card form factor is an excellent choice for enterprises that need more AI computing power than CPUs can provide but aren’t ready to invest in dedicated GPU accelerator platforms. Available in air-cooled systems with up to eight accelerator cards, AMD Instinct MI350P PCIe cards are ideal for small, medium and large AI models for inference and RAG pipelines.

Don’t Just Scale AI. Scale ROI.

AMD Instinct MI350P PCIe cards are engineered to deliver exceptional AI performance with excellent cost and leadership performance. Key features help increase performance, simplify deployment and reduce costs so you can move from evaluation to real outcomes:

• Native support for lower-precision MXFP6 and MXFP4, which deliver high throughput.
• Acceleration through sparsity support for most mainstream 8- and 16-bit precisions.
• Estimated 2,299 teraflops (TFLOPS) and up to 4,600 peak TFLOPS at MXFP4, the highest performance currently available in an enterprise PCIe card.
• Estimated 144GB of high bandwidth memory 3e (HBM3E) running at up to 4TB/s.
• Open ecosystem with low- and no-cost development stack options simplifies deployment and helps lower operating expenses.

Enterprise AI Software – Develop with Your AI Stack. Your Way, Today.

We built AMD Instinct MI350P PCIe cards with open standards for cross-platform interoperability. Their addition continues our strategy of enabling a fully open AI ecosystem and providing customer choice in enterprise environments.

Think of the AMD enterprise AI stack as a foundational component, integrating seamlessly with a broad ecosystem of AI software and tools. It includes the Kubernetes GPU Operator for full life cycle management, cloud-native AMD Inference Microservices and native support for AI frameworks such as PyTorch. All this enables you to migrate inference workloads with minimal code changes.

We provide the open-source AMD enterprise AI reference stack to our partners at no licensing cost. It offers greater code transparency and helps reduce operating expenses. When combined with AMD Instinct MI350P PCIe cards and partner-delivered solutions, the stack enables your organizations to get up and running quickly on-premises without ongoing per-token charges.

Native Acceleration for Enterprise AI Precision Levels

AMD Instinct MI350P PCIe cards support the spectrum of precision levels that enterprise AI models rely on most.

While lower-precision MXFP6 and MXFP4 offer maximized performance in pure TFLOPS and efficient model implementations, higher precision formats, like INT8 and BF16, benefit from the sparsity support on the AMD Instinct MI350P GPU to deliver efficient performance. Regardless of the precision, enterprises will find that AMD Instinct MI350P PCIe cards are designed to deliver maximum GPU throughput and reduced memory usage to help lower power and cooling demands.

Support for FP8, MXFP8 and MXFP4 is a major reason AMD Instinct MI350P PCIe cards can process today’s AI workloads within standard, air-cooled data centers.

Deploy Enterprise AI Where You are Today

With AMD Instinct MI350P PCIe cards, your enterprise can move quickly from bare-metal infrastructure to production-ready AI systems on a strong foundation. They enable you to migrate workloads without code rewrites, integrate with existing AI pipelines and scale with evolving workloads.

Adopting AI doesn’t mean rebuilding infrastructure from the ground up. With AMD Instinct MI350P PCIe cards, enterprises can run more models and serve more users within their existing data centers.