AMD Unveils New Era of Silicon at CES 2026, From Powering the Future PC to the World’s Biggest Data Centers
LAS VEGAS—AMD arrived at CES 2026 this year with one clear message: the future of computing, both personal and massive, runs on accelerated hardware. The company’s keynote presentation, led by CEO Dr. Lisa Su, was a dazzling showcase of new silicon, offering a deep dive into the next-generation of Ryzen processors for everyday PCs and a significant roadmap update for its high-powered data center chips.
The biggest news for consumers centers on the new Ryzen AI 400 Series of mobile processors. These chips are set to power the next wave of ultra-thin laptops coming in 2026. Built on the ‘Zen 5’ CPU cores and featuring an upgraded XDNA 2 Neural Processing Unit (NPU), the flagship chips in the series can deliver up to 60 Trillion Operations Per Second (TOPS) of acceleration. This is a critical performance leap, moving the mobile processor firmly into the territory required for advanced local computing applications, putting it in a strong competitive position against rivals like Intel and Qualcomm.
AMD also unveiled the Ryzen AI Max+ Series, including the new 392 and 388 SKUs. These are designed for premium ultra-thin notebooks and small form factor desktops, incorporating a powerful integrated graphics engine along with the updated NPU to handle demanding creative and gaming workloads in a highly efficient design. For desktop enthusiasts, the company introduced the highly anticipated Ryzen 7 9850X3D. This new gaming processor is based on the ‘Zen 5’ architecture and utilizes the company’s signature 3D V-Cache technology, promising a major speed boost for gamers.
Moving from the desktop to the data center, AMD signaled a profound acceleration in its high-end computing strategy with the introduction of its new rack-scale platform, codenamed Helios. This mammoth system is designed to be the blueprint for the next generation of cloud-scale infrastructure. At its core, the Helios rack features the next-generation EPYC “Venice” CPU, built on the “Zen 6” architecture, paired with the powerful new **Instinct MI455X GPU** accelerators.
The MI455X GPUs are groundbreaking, marking the company’s first high-performance chips produced using a 2-nanometer class manufacturing process. A single Helios rack integrates 72 of these accelerators, delivering immense computational performance, with a capability of up to 2.9 exaFLOPS for large-scale inferencing tasks. Adding weight to the announcement, OpenAI President Greg Brockman made a surprise appearance during the keynote, underscoring the collaboration between the two companies on open systems development and next-generation large-scale training.
Looking further ahead, AMD offered a tantalizing glimpse into its 2027 roadmap with a preview of the **Instinct MI500 Series** GPUs. Built on the CDNA 6 architecture and paired with high-bandwidth memory, the MI500 is projected to deliver a remarkable 1,000 times increase in acceleration performance compared to the previous generation of Instinct chips, a clear and aggressive target set against the current market leader in the data center space. From the notebook in your hand to the servers powering the largest digital services in the world, AMD is clearly betting big on its accelerated future.