Core Viewpoint - Nanwei Semiconductor (NVTS.US) shares rose over 16% to $17.1 following the announcement of new power semiconductor products designed for NVIDIA's (NVDA.US) 800VDC AI factory power architecture [1] Group 1: Product Launch - The newly launched products include 100V GaN FETs, 650V GaN devices, and high-voltage SiC (silicon carbide) products [1] - These products aim to significantly enhance energy efficiency, power density, and operational performance, providing groundbreaking power solutions for AI data centers [1] Group 2: Technology and Application - The NVIDIA 800VDC architecture is a new power distribution system tailored for AI factories, capable of supporting large-scale, high-performance AI workloads [1] - This architecture directly powers IT cabinets to drive AI infrastructure, including the Rubin Ultra [1] Group 3: Product Advantages - The 100V GaN FET products demonstrate clear advantages in efficiency, power density, and thermal management [1] - The 650V GaN combination integrates GaNSafe power ICs that consolidate driving, control, sensing, and protection functions, facilitating NVIDIA's transition from traditional architectures to 800VDC [1]

Core Viewpoint - The article highlights the significant collaboration between InnoScience and NVIDIA, focusing on the implementation of the 800VDC power architecture in AI data centers, which is expected to enhance efficiency and performance in AI computing [3]. Group 1: Collaboration and Market Impact - InnoScience's stock surged by over 60% during trading on August 1, reaching a new high since its listing, closing up 30.91% at HKD 57.60 per share following the announcement of the partnership with NVIDIA [1][3]. - The collaboration aims to promote the large-scale deployment of the 800VDC power architecture, which is designed for high-efficiency power supply in megawatt-level computing infrastructures [3]. Group 2: Technical Advantages of 800VDC Architecture - The 800VDC architecture offers significant advantages over traditional 54V systems, including improved system efficiency, reduced thermal losses, and enhanced reliability, enabling a 100 to 1000 times increase in AI computing power [3][5]. - InnoScience's third-generation GaN devices provide a comprehensive power solution from 800V input to GPU terminals, covering a voltage range from 15V to 1200V, facilitating the integration of GaN technology with NVIDIA's architecture [3]. Group 3: Future Prospects and Cost Efficiency - The 800V HVDC architecture is positioned as a future-proof solution for AI infrastructure, with full-scale production expected by 2027, ensuring scalability for increasingly demanding AI workloads [5]. - The implementation of this architecture could reduce the total cost of ownership (TCO) by up to 30% through improved efficiency and system architecture enhancements [5].