AMD Unveils Instinct AI GPU Roadmap, Introduces MI350X CDNA 4, and 1-year Cadence

Computex Taipei

At Computex 2024, AMD made a splash with its AMD Instinct accelerator family, demonstrating its continued momentum in the AI space. Dr. Lisa Su, Chair and CEO of AMD, took the stage for the opening keynote and unveiled an expanded roadmap for the AMD Instinct accelerators, promising a yearly release of cutting-edge AI performance and memory capabilities.

The first addition to the updated roadmap is the AMD Instinct MI325X accelerator, set to hit the market in Q4 2024. This powerful accelerator will be followed by the AMD Instinct MI350 series, expected in 2025, which will leverage the new AMD CDNA 4 architecture to deliver up to a 35x increase in AI inference performance compared to the previous generation.

Looking ahead to 2026, AMD plans to release the AMD Instinct MI400 series, based on the highly anticipated AMD CDNA “Next” architecture. This release aims to push the boundaries of performance and efficiency for both AI inference and large-scale AI training.

Brad McCredie, corporate vice president of Data Center Accelerated Compute at AMD, highlighted the strong adoption of the AMD Instinct MI300X accelerators by major partners and customers such as Microsoft Azure, Meta, Dell Technologies, HPE, and Lenovo. This widespread adoption is a testament to the exceptional performance and value proposition offered by these accelerators.

In addition to hardware advancements, AMD also showcased the maturity of its AI software ecosystem. The AMD ROCm 6 open software stack continues to evolve, enabling impressive performance gains for popular AI models. For example, using eight AMD Instinct MI300X accelerators with ROCm 6 running Meta Llama-3 70B, customers can achieve 1.3x better inference performance and token generation compared to the competition.

AMD’s collaboration with Hugging Face, the leading repository for AI models, is also worth noting. Hugging Face is now testing 700,000 of their most popular models nightly to ensure seamless compatibility with AMD Instinct MI300X accelerators. Furthermore, AMD remains committed to upstream contributions to popular AI frameworks like PyTorch, TensorFlow, and JAX.

To meet the growing demand for AI compute, AMD revealed an updated annual cadence for the AMD Instinct accelerator roadmap. This strategic move aims to fuel the development of next-generation AI models. The roadmap includes exciting additions such as the AMD Instinct MI325X accelerator, boasting an impressive 288 GB of HBM3E memory and 6 terabytes per second of memory bandwidth. This accelerator will also offer industry-leading compute performance, outperforming the competition by 1.3x.

The future looks even brighter with the upcoming AMD Instinct MI350X accelerator, which will leverage the advanced 3 nm process technology and support the FP4 and FP6 AI datatypes. Furthermore, the AMD CDNA “Next” architecture, expected in 2026, promises to unlock additional performance and efficiency for inference and large-scale AI training.

AMD’s commitment to innovation has not gone unnoticed, as partners and customers across various industries continue to embrace the AMD Instinct MI300X accelerators. Microsoft Azure is utilizing these accelerators for its Azure OpenAI services and Azure ND MI300X V5 virtual machines. Dell Technologies has incorporated the MI300X accelerators into its PowerEdge XE9680 for enterprise AI workloads. Supermicro is offering multiple solutions featuring AMD Instinct accelerators, while Lenovo is powering Hybrid AI innovation with the ThinkSystem SR685a V3. HPE is also leveraging these accelerators to accelerate AI workloads in the HPE Cray XD675.

With its relentless pace of innovation, AMD is poised to drive the next evolution of data center AI training and inference. The company’s dedication to delivering leadership capabilities and performance aligns with the expectations of the AI industry and its customers. As the AMD Instinct accelerator family continues to grow, it solidifies AMD’s position as a key player in the AI market.

Background Information

---

About AMD: AMD, a large player in the semiconductor industry is known for its powerful processors and graphic solutions, AMD has consistently pushed the boundaries of performance, efficiency, and user experience. With a customer-centric approach, the company has cultivated a reputation for delivering high-performance solutions that cater to the needs of gamers, professionals, and general users. AMD's Ryzen series of processors have redefined the landscape of desktop and laptop computing, offering impressive multi-core performance and competitive pricing that has challenged the dominance of its competitors. Complementing its processor prowess, AMD's Radeon graphics cards have also earned accolades for their efficiency and exceptional graphical capabilities, making them a favored choice among gamers and content creators. The company's commitment to innovation and cutting-edge technology continues to shape the client computing landscape, providing users with powerful tools to fuel their digital endeavors.

  

---

About Dell: Dell is a globally renowned technology leader providing comprehensive solutions in the field of hardware, software, and services. Renowned for its customizable computers and enterprise solutions, Dell offers a diverse range of laptops, desktops, servers, and networking equipment. With a commitment to innovation and customer satisfaction, Dell caters to a wide range of consumer and business needs, making it a prominent player in the tech industry.

  

---

About Lenovo: Lenovo, formerly known as "Legend Holdings," is a prominent global technology company that offers an extensive portfolio of computers, smartphones, servers, and electronic devices. Notably, Lenovo acquired IBM's personal computer division, including the ThinkPad line of laptops, in 2005. With a strong presence in laptops and PCs, Lenovo's products cater to a wide range of consumer and business needs. Committed to innovation and quality, Lenovo delivers reliable and high-performance solutions, making it a significant player in the tech industry.

  

---

About Microsoft: Microsoft, founded by Bill Gates and Paul Allen in 1975 in Redmond, Washington, USA, is a technology giant known for its wide range of software products, including the Windows operating system, Office productivity suite, and cloud services like Azure. Microsoft also manufactures hardware, such as the Surface line of laptops and tablets, Xbox gaming consoles, and accessories.

  

---

About Supermicro: Supermicro is a reputable American technology company founded in 1993 and headquartered in San Jose, California. Specializing in high-performance server and storage solutions, Supermicro has become a trusted name in the data center industry. The company offers a wide range of innovative and customizable server hardware, including motherboards, servers, storage systems, and networking equipment, catering to the needs of enterprise clients, cloud service providers, and businesses seeking reliable infrastructure solutions.

  

Event Info

---

About Computex: Computex, held annually in Taipei, Taiwan, stands as one of the world's leading technology trade shows, showcasing cutting-edge innovations in computing hardware, software, and emerging technologies. With a focus on industry trends and product launches, it serves as a pivotal platform for tech giants and startups alike to unveil their latest advancements and forge key partnerships, attracting a global audience of industry professionals, enthusiasts, and media representatives.

Technology Explained

---

HBM3E: HBM3E is the latest generation of high-bandwidth memory (HBM), a type of DRAM that is designed for artificial intelligence (AI) applications. HBM3E offers faster data transfer rates, higher density, and lower power consumption than previous HBM versions. HBM3E is developed by SK Hynix, a South Korean chipmaker, and is expected to enter mass production in 2024. HBM3E can achieve a speed of 1.15 TB/s and a capacity of 64 GB per stack. HBM3E is suitable for AI systems that require large amounts of data processing, such as deep learning, machine learning, and computer vision.