Samsung and South Korean search giant NAVER have reached a partnership to develop dedicated chips for large-scale AI models

CTOnews.com, December 6 (Xinhua)-- Samsung has announced a partnership with South Korean search giant NAVER to develop special chips for large-scale AI models. The two sides said that the amount of data of artificial intelligence has increased exponentially in recent years, and that the current technology has limitations in terms of performance and efficiency, unable to properly handle such a large workload.

CTOnews.com learned that Samsung and NAVER decided to work together to create semiconductor chips and other solutions optimized for AI. In this cooperation, with the help of Samsung's computing storage (SmartSSD), CXL (Compute Express Link), PIM (Processor In Memory) and PNM (Processing Near Memory) technologies, combined with NAVER's rich experience in the field of AI algorithms, the combination of software and hardware will greatly improve the artificial intelligence workload.

Samsung Electronics will contribute its hardware and manufacturing expertise, while NAVER will bring its expertise in the development and verification of AI-based algorithms. NAVER will improve HyperCLOVA (a very large-scale model with more than 200 billion parameters) and its compression algorithm to improve efficiency.