AMD and Samsung forge strategic partnership for AI memory supply with foundry deal on the table

AMD and Samsung ink memory supply memorandum for EPYC and Instinct products -- unprecedented deal also includes scope for foundry partnership

AMD and Samsung this week signed a rather unprecedented memorandum of understanding (MOU) that involves the strategic supply of memory for AMD's next-generation EPYC and Instinct MI455X products, as well as discussions of a potential foundry partnership. The MOU aims to ensure that AMD gets enough memory for its next-generation CPU and AI accelerator products throughout their life span, whereas Samsung will get a chance to serve as a foundry for a leading supplier of CPUs, AI GPUs, and other products. Samsung has been the primary supplier of HBM3E for AMD's Instinct MI350X and MI355X AI accelerators, so it is not surprising that the core of the new agreement is Samsung's role as a primary supplier of HBM4 memory for AMD's upcoming Instinct MI455X accelerator. As demand for all types of memory in general and HBM4 in particular exceeds supply, AMD must ensure that its next-generation MI455X accelerator (and possibly other Instinct MI400-series products) gets enough memory. AMD's Instinct MI455X is projected to use 12 HBM4 12-Hi memory stacks for a total of 432 GB. Given such demands, it makes great sense for AMD to sign a strategic memory supply deal, though for now the two companies limited themselves to the MOU. Meanwhile, Samsung said that its HBM4 memory stacks are based on memory devices produced using its 1c (6 Generation 10nm-class) process technology and use a base die made on a 4nm-class logic process, which enables them to achieve a rather unprecedented data transfer rate of 13 GT/s (thus provide up to 3.3 TB/s of bandwidth per stack), which is well above JEDEC's recommended 8 GT/s for HBM4. Yet, memory speeds supported by Instinct MI455X are considerably slower. In addition to HBM4, the MOU between the two companies also includes DDR5 supply for AMD's 6 Generation EPYC processors codenamed 'Venice,' as well as AMD's Helios rack-scale systems for AI data centers that rely on EPYC 'Venice' and Instinct MI455X. Again, the move aims to ensure a steady memory supply for next-generation AI systems running AMD's CPUs and AI accelerators to ensure a competitive position against Nvidia's Rubin-powered machines. Perhaps the most intriguing part of the announcement is that the MOU can open the door for a potential foundry relationship, under which Samsung could manufacture future AMD products. While no specific nodes or chips were disclosed, this signals a willingness to expand cooperation into logic production alongside memory supply. To put this into context, AMD has been drifting away from GlobalFoundries since late 2018, and by now virtually all of its advanced products are made at TSMC. Shifting certain advanced production to Samsung is costly, though if this ensures that the South Korean company will provide much-needed memory to support AMD's AI efforts, then it looks like the developer of EPYC CPUs and Instinct accelerators may be willing to adopt a dual-source foundry strategy and make at least some of its products at Samsung Foundry. "Samsung and AMD share a commitment to advancing AI computing, and this agreement reflects the growing scope of our collaboration," said Young Hyun Jun, Vice Chairman & CEO of Samsung Electronics. "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." Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

Samsung and AMD deepen AI memory pact with potential foundry deal on the table

Serving tech enthusiasts for over 25 years. TechSpot means tech analysis and advice you can trust. What we know so far: Samsung Electronics and AMD are strengthening their partnership to secure a foothold in the fast-evolving artificial intelligence hardware market. This week, the two companies signed a memorandum of understanding that expands their collaboration beyond memory supply to include potential chip manufacturing, highlighting how strategic alliances are reshaping global semiconductor competition. The agreement positions Samsung as a key supplier for AMD's next-generation AI products. Under the terms, Samsung will provide its forthcoming high-bandwidth memory, HBM4, for AMD's Instinct MI455X accelerators - processors designed specifically for AI workloads. Samsung will also deliver optimized DDR5 memory for AMD's upcoming sixth-generation Epyc server chips, a move expected to bolster AMD's position against Nvidia and Intel in the data center market. What makes the deal particularly notable is its potential expansion beyond memory. The two companies said they plan to explore a foundry partnership that could see Samsung fabricating AMD's future chips. While no formal manufacturing contract has been announced, these discussions signal AMD's continued interest in diversifying production away from a single source. For Samsung, the collaboration offers a stronger role in the AI memory supply chain at a time when demand for HBM chips is outpacing production. The South Korean company already provides AMD with HBM3E chips used in current MI350X and MI355X accelerators. Becoming the key supplier for the new HBM4 generation would further solidify that relationship, even as major competitors ramp up their own AI offerings. The timing of the agreement is notable. It was announced the same week that Nvidia CEO Jensen Huang used the company's annual GTC event to showcase new AI chips and confirm a separate foundry partnership with Samsung. Nvidia's endorsement of Samsung's HBM4 technology drew attention to the memory maker's progress in a market long dominated by SK Hynix. According to Counterpoint Research, SK Hynix holds roughly 57% of the high-bandwidth memory market - more than double Samsung's estimated 22%. For AMD, securing memory supply is increasingly critical. The chipmaker has recently landed large, long-term contracts to provide AI accelerators to major data center players. In February, it agreed to sell up to $60 billion in AI chips to Meta Platforms over five years in a deal that gives Meta the option to acquire up to a 10% stake in AMD. AMD reached a similar deal with OpenAI in late 2025. The Samsung partnership reinforces that trend. As more companies compete for limited HBM capacity, alliances between chip designers and memory specialists are becoming as much a competitive strategy as a logistical necessity. For Samsung, the agreement not only helps it gain ground on its biggest rival but also underscores its ambitions to be integral to the AI computing ecosystem, rather than merely a supplier.

AMD Finally Secures Samsung's HBM4 Supply Following NVIDIA's Footsteps, but the Deal Comes With a Surprising Caveat

AMD's CEO, Lisa Su, and her visit to South Korea were apparently aimed at securing HBM4 supply from Samsung, and it appears she has succeeded in securing the desired capacity. The AI industry, particularly companies involved in the supply chain, has become a much more important entity for manufacturers like AMD, as supply constraints have given them the upper hand in negotiations. A few days ago, we discussed how AMD's Lisa Su plans to visit South Korea to meet with Samsung executives and other partners, and the meeting took place today. According to a Samsung announcement, AMD has secured HBM4 supply for its Instinct MI455X AI accelerators, and will also work on DDR5 for Venice CPUs, but the partnership has an interesting caveat. Under the MOU, Samsung and AMD will align on primary HBM4 supply for the next-generation AMD AI accelerator, the AMD Instinct MI455X GPU, as well as advanced DRAM solutions for 6th Gen AMD EPYC CPUs, codenamed "Venice." These technologies will support next-generation AI systems combining AMD Instinct GPUs, AMD EPYC CPUs and rack-scale architectures such as the AMD Helios platform. - Samsung The tables have turned significantly for Samsung in the past few months, as it has seen strong momentum in its HBM business. Following breakthroughs with HBM3 and HBM3E, Samsung's HBM4 solution became highly sought after by the likes of NVIDIA, mainly because it features pin speeds of up to 13 Gbps, which are among the highest available. Team Green was early to secure HBM4 capacity from Samsung for Vera Rubin, giving it an edge over AMD as well. And for Samsung, well, they were in a superior position when negotiating with Lisa Su, which is why the cooperation isn't just limited to memory. The Korean media outlet Chosun Biz reports that the Korean giant has kept the conditions for producing some of AMD's "advanced" AI chips at Samsung Foundry, and while the details of the potential collaboration aren't certain, this does seem like a wise move. We do know that Samsung Foundry has witnessed significant momentum in recent times, following deals with NVIDIA, Tesla, Apple, and other customers, and by adding AMD to its portfolio. This essentially positions the foundry division as a reliable entity to meet semiconductor demand. There were rumors that Samsung was looking to produce AMD's EPYC Venice CPUs using its SF2 process, but the Korean report mentions "advanced" AI chips, suggesting accelerators.

Samsung and AMD Expand Strategic Partnership for Next Generation AI Memory

The signing ceremony was held at Samsung's most advanced chip manufacturing complex in Pyeongtaek, Korea, attended by Dr. Lisa Su, Chair and CEO of AMD, and Young Hyun Jun, Vice Chairman & CEO of Samsung Electronics. "Samsung and AMD share a commitment to advancing AI computing, and this agreement reflects the growing scope of our collaboration," said Young Hyun Jun, Vice Chairman & CEO of Samsung Electronics. "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."

Samsung signs MOU with AMD for next-gen AI Memory

Samsung and AMD have announced the signing of a Memorandum of Understanding (MOU) to broaden their strategic partnership concerning next-generation artificial intelligence (AI) memory and computing technologies. The agreement was formally signed at Samsung's semiconductor manufacturing complex in Pyeongtaek, South Korea. The ceremony was attended by Dr. Lisa Su, Chair and CEO of AMD, and Young Hyun Jun, Vice Chairman and CEO of Samsung Electronics. Under the terms of the MOU, Samsung will serve as the primary supplier of HBM4 memory for AMD's upcoming AI accelerator, the AMD Instinct MI455X GPU. The collaboration also includes the development of advanced DRAM solutions -- specifically high-performance DDR5 memory -- optimized for the 6th Generation AMD EPYC CPUs, internally codenamed "Venice." These combined memory and processing technologies are designed to support next-generation AI systems, including rack-scale architectures like the AMD Helios platform. The companies noted that the collaboration aims to address the increasing requirements for memory bandwidth and power efficiency in modern AI and data center workloads. Samsung's HBM4 memory, which has entered mass production, is manufactured using the company's 6th-generation 10-nanometer-class (1c) DRAM process and features a 4nm logic base die. According to Samsung, the HBM4 memory provides processing speeds of up to 13 gigabits-per-second (Gbps) and a maximum bandwidth of 3.3 terabytes-per-second (TB/s). The AMD Instinct MI455X GPU, utilizing this HBM4 memory, is positioned to handle high-performance tasks such as AI model training and inference. The GPU will function as a core component of the AMD Helios rack-scale architecture, which focuses on performance scalability for AI infrastructure. In addition to memory supply and optimization, the MOU indicates that Samsung and AMD will explore opportunities for a foundry partnership. Under such an arrangement, Samsung would potentially provide semiconductor manufacturing services for future AMD products. This recent agreement extends a nearly two-decade relationship between Samsung and AMD, which has spanned mobile, graphics, and computing technologies. Most recently, Samsung served as the primary HBM3E memory supplier for AMD's current generation of AI accelerators, the Instinct MI350X and MI355X. Commenting on this, Young Hyun Jun, Vice Chairman & CEO of Samsung Electronics, 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, Chair and CEO of AMD, said: