AMD Introduces Its Open, Scalable Rack-Scale AI Infrastructure Built on Industry Standards
AMD recently unveiled its comprehensive, end-to-end integrated AI platform vision at the 2025 Advancing AI event. The company introduced its open, scalable rack-scale AI infrastructure that is built on industry standards, showcasing its commitment to driving innovation in the AI space. Partnering with industry leaders, AMD demonstrated how they are shaping the future of AI with cutting-edge technologies and solutions.
AMD's Leadership in AI Solutions
AMD announced a range of hardware, software, and solutions designed to power various AI applications. The Instinct MI350 Series GPUs, including the MI350X and MI355X models, set new benchmarks for performance, efficiency, and scalability in generative AI and high-performance computing. These GPUs deliver significant improvements in AI compute and inferencing capabilities, enabling transformative AI solutions across industries.
Furthermore, AMD showcased its open-standards rack-scale AI infrastructure, already deployed with Instinct MI350 Series accelerators, 5th Gen AMD EPYC processors, and AMD Pensando Pollara NICs in hyperscaler deployments like Oracle Cloud Infrastructure (OCI). This infrastructure is set to be broadly available in the second half of 2025, offering enhanced AI performance and efficiency.
Next-Generation AI Innovation
AMD also provided a glimpse into its future with the preview of the "Helios" AI rack. This next-generation AI rack will be powered by the upcoming AMD Instinct MI400 Series GPUs, expected to deliver significant performance improvements for running inference on Mixture of Experts models. Coupled with Zen 6-based AMD EPYC "Venice" CPUs and AMD Pensando "Vulcano" NICs, the Helios rack promises to push the boundaries of AI processing.
Empowering the AI Community
To further support AI development, AMD announced the broad availability of the AMD Developer Cloud, providing developers and open-source communities with a fully managed cloud environment for high-performance AI projects. By offering access to ROCm 7 and the AMD Developer Cloud, AMD aims to democratize access to next-gen compute resources and foster collaboration within the AI community.
Collaboration and Innovation
AMD's strategic collaborations with industry leaders like Meta, OpenAI, Microsoft, and others showcased the transformative power of AMD's AI solutions. From model building to inference deployment, these collaborations highlight the diverse applications of AMD's AI technologies across various sectors, driving innovation and progress in the AI landscape.
Overall, AMD's advancements in open, scalable rack-scale AI infrastructure underscore the company's commitment to advancing AI technologies and empowering the global AI community with cutting-edge solutions.
