Samsung Electronics and NAVER Partner to Develop Large-Scale AI-Optimized Semiconductor Solutions – Samsung Global Newsroom

Both companies intend to combine their prowess in semiconductor design and manufacturing with proven AI capabilities to maximize the speed and energy efficiency of large-scale AI models.

Samsung Electronics, the world leader in advanced memory technologies, and NAVER Corporation, a global Internet company with cutting-edge AI technology, today announced a far-reaching collaboration to develop suitable semiconductor solutions to large-scale artificial intelligence (AI) models. Leveraging Samsung’s next-generation memory technologies such as Compute Storage, In-Memory Processing (PIM) and Near-Memory Processing (PNM), as well as Compute Express Link (CXL), the companies intend to pool their hardware and software resources to dramatically accelerate the handling of massive AI workloads.

Recent advances in large-scale AI have led to exponential growth in the volumes of data to be processed. However, the performance and efficiency limitations of today’s computing systems pose significant challenges in meeting these heavy computational demands, fueling the need for new AI-optimized semiconductor solutions.

The development of such solutions requires a strong convergence of semiconductor and AI disciplines. Samsung combines its expertise in semiconductor design and manufacturing with NAVER’s experience in developing and verifying AI algorithms and AI-based services, to create solutions that drive performance and the energy efficiency of large-scale AI at a higher level.

For years, Samsung has been introducing memory and storage that supports high-speed data processing in AI applications, computing storage (SmartSSD) and PIM-enabled high-bandwidth memory (HBM-PIM ) to next-generation memory supporting Compute Express Link (CXL). Samsung will now join NAVER in optimizing these memory technologies to advance large-scale AI systems.

NAVER will continue to refine HyperCLOVA, a hyperscale language model with over 200 billion parameters, while improving its compression algorithms to create a more simplified model that dramatically increases computational efficiency.

“Through our collaboration with NAVER, we will develop cutting-edge semiconductor solutions to solve the memory bottleneck in large-scale AI systems,” said Jinman Han, Executive Vice President of Sales and global memory marketing at Samsung Electronics. “With tailored solutions that reflect the most pressing needs of AI service providers and users, we are committed to expanding our market-leading memory portfolio, including computing storage, PIM and more, to fully adapt to the ever-increasing scale of data.”

“By combining our knowledge and know-how gained from HyperCLOVA with Samsung’s prowess in semiconductor manufacturing, we believe we can create a whole new class of solutions that can better meet the challenges of artificial intelligence of today,” said Suk Geun Chung, head of NAVER CLOVA CIC. . “We look forward to expanding our AI capabilities and strengthening our AI competitive advantage through this strategic partnership.”