During the CES 2026 keynote in Las Vegas, AMD outlined what CEO Lisa Su framed as the next phase of AI adoption, spanning data center infrastructure, AI PCs, and embedded edge systems, with partners including OpenAI, Luma AI, Liquid AI, World Labs, Blue Origin, Generative Bionics, AstraZeneca, Absci, and Illumina highlighting use cases powered by AMD hardware and software.
At the center of AMD’s infrastructure message was an early look at “Helios,” a rack-scale platform positioned as a blueprint for yotta-scale AI compute. AMD said Helios is designed around an open, modular rack architecture that can evolve across generations and connect large numbers of accelerators with high-speed networking to support trillion-parameter training, targeting up to 3 AI exaflops of performance in a single rack. The platform is built on AMD Instinct MI455X accelerators, AMD EPYC “Venice” CPUs, and AMD Pensando “Vulcano” network interface cards, unified through the AMD ROCm software ecosystem.
AMD also expanded its Instinct accelerator narrative with the introduction of the Instinct MI440X, a GPU aimed at on-premises enterprise deployments in an eight-GPU form factor intended to fit within existing infrastructure, supporting training, fine-tuning, and inference. The company also previewed the next-generation Instinct MI500 Series, planned for launch in 2027, and said the family is on track to deliver up to a 1,000x increase in AI performance compared with the Instinct MI300X introduced in 2023, driven by CDNA 6, a 2nm process, and HBM4E memory. The keynote also referenced the previously announced MI430X GPUs for scientific, HPC, and sovereign AI use cases, including planned deployments in systems such as Discovery at Oak Ridge National Laboratory and France’s Alice Recoque exascale supercomputer.
On the client side, AMD introduced new Ryzen AI platforms to broaden its AI PC portfolio and deepen developer support. The company said its Ryzen AI 400 Series and Ryzen AI PRO 400 Series platforms deliver a 60 TOPS NPU and include ROCm platform support for cloud-to-client AI scaling, with first systems expected to ship in January 2026 and broader OEM availability targeted for the first quarter of 2026. AMD also highlighted Ryzen AI Max+ configurations, including the Ryzen AI Max+ 392 and 388, positioning them for advanced local inference and creator workflows with support for models up to 128 billion parameters alongside 128GB of unified memory.
For developers, AMD unveiled the Ryzen AI Halo Developer Platform, described as a compact small form factor desktop designed to bring high-performance AI development capabilities to a broader audience using Ryzen AI Max+ Series processors, with availability expected in the second quarter of 2026.
AMD also pushed deeper into edge and “physical AI” with the introduction of Ryzen AI Embedded processors, a new portfolio of embedded x86 chips intended for AI-driven applications in constrained environments. The company said the new P100 and X100 Series processors are designed for use cases ranging from automotive digital cockpits and smart healthcare to autonomous systems and humanoid robotics.
The keynote included a U.S. policy and research angle, with White House Office of Science and Technology Policy Director Michael Kratsios joining Su on stage to discuss AMD’s role in the Genesis Mission, a public-private initiative focused on U.S. leadership in AI technologies. AMD linked the effort to two AMD-powered AI supercomputers at Oak Ridge National Laboratory, Lux and Discovery. AMD also announced a $150 million commitment aimed at expanding access to AI education in classrooms and communities, and closed by recognizing participants in the AMD AI Robotics Hackathon held in partnership with Hack Club.
KEY QUOTE:
“At CES, our partners joined us to show what’s possible when the industry comes together to bring AI everywhere, for everyone. As AI adoption accelerates, we are entering the era of yotta-scale computing, driven by unprecedented growth in both training and inference. AMD is building the compute foundation for this next phase of AI through end-to-end technology leadership, open platforms, and deep co-innovation with partners across the ecosystem.”
Dr. Lisa Su, Chair And CEO, AMD
