AMD is buying Samsung’s HBM4 high-bandwidth memory for its upcoming “AI accelerator”, the Instinct MI455X GPU, as well as 6th gen AMD Epyc CPUs codenamed Venice. AI data center operators can combine Instinct GPUs with Epyc CPUs and rack-scale architectures such as the AMD Helios platform to support next-gen AI systems.
The memorandum of understanding between Samsung and AMD on these matters was announced today at Samsung’s most advanced chip manufacturing complex in Pyeongtaek, Korea. The signing ceremony was attended by AMD CEO Dr. Lisa Su and Samsung Electronics CEO and Vice Chairman Young Hyun Jun.
Young Hyun Jun said:
Samsung and AMD share a commitment to advancing AI computing, and this agreement reflects the growing scope of our collaboration. From industry-leading HBM4 and next-generation memory architectures to cutting-edge foundry and advanced packaging, Samsung is uniquely positioned to deliver unrivaled turnkey capabilities that support AMD’s evolving AI roadmap.
Dr. Lisa Su said:
Powering the next generation of AI infrastructure requires deep collaboration across the industry. We are thrilled to expand our work with Samsung, bringing together their leadership in advanced memory with our Instinct GPUs, Epyc CPUs and rack-scale platforms. Integration across the full computing stack, from silicon to system to rack, is essential to accelerating AI innovation that translates into real-world impact at scale.
Samsung’s HBM4 is made on a 10nm process and a 4nm logic base die, and features processing speeds of up to 13Gbps and a maximum bandwidth of 3.3TBps. Samsung and AMD will also work together on high-performance DDR5 memory optimized for the 6th gen Epyc CPUs. The two companies will also “discuss opportunities for foundry partnership”, the official press release says, “through which Samsung would provide foundry services for next-generation AMD products”.