AMD introduces the Ryzen AI Halo, a new “agent computer” platform designed to simplify and accelerate local AI development with a unified developer center, continuous support, and partnerships like Hugging Face to optimize AI models. Alongside this, AMD unveils the Ryzen AI Max 400 series hardware, featuring advanced architecture and large unified memory to enable complex, real-time AI workloads on personal devices, marking a significant step toward accessible and efficient local AI computing.
The video announces two major developments from AMD that aim to transform how developers build and deploy AI systems locally. AMD emphasizes that AI is no longer experimental but is being deployed in real-world applications, with increasing demands for performance, memory, and efficiency. To address these challenges, AMD introduces the Ryzen AI Halo, a new category of “agent computer” designed to enable developers to go from concept to working AI system quickly and efficiently. The Ryzen AI Halo is built around four core principles: speed of productivity, simplicity of setup, cost efficiency, and continuous developer support.
Central to the Ryzen AI Halo experience is the Ryzen AI developer center, which comes pre-installed and optimized to eliminate setup friction and dependency issues. This unified platform offers continuously updated and validated tools, frameworks, and models, ensuring developers can confidently build AI applications that run effectively from day one. AMD also highlights its AI developer program, which provides direct access to support and guidance, reinforcing its commitment to helping developers succeed in a rapidly evolving AI landscape.
AMD has partnered with Hugging Face, a leading AI community, to enhance local AI development. Clem Delangue, CEO of Hugging Face, praises the local-first approach of the Ryzen AI Halo, which empowers developers to build AI solutions sustainably, privately, and independently of cloud services. Hugging Face is collaborating with AMD to optimize popular AI models and integrate native Ryzen AI support, enabling access to millions of open-source models and datasets for faster and more specialized AI applications.
Looking ahead, AMD unveils the Ryzen AI Max 400 series, set to launch on Ryzen AI Halo platforms in the third quarter. Powered by a new silicon architecture called Gorgam Halo, this series offers up to 192 GB of unified memory and 160 GB dedicated to AI processing. This hardware advancement supports more complex workloads, including running multiple AI agents in parallel and handling models with over 300 billion parameters locally. This marks a significant leap in enabling real-time, intelligent, and adaptive AI systems on personal devices.
In conclusion, AMD envisions a future where local AI development is unrestricted by infrastructure limitations, allowing ideas to progress at the speed of computation and making AI intelligence personal and accessible. The company commits to ongoing innovation and collaboration with the developer community to push the boundaries of AI performance and capability, signaling that this is just the beginning of a new era in local AI computing.