Group 1 - Southbound funds recorded a net sell of HKD 18.092 billion in Hong Kong stocks on August 4, marking the largest single-day net sell since May 12 [1] - Notable net purchases included Alibaba-W at HKD 728 million, InnoCare Pharma at HKD 362 million, Kuaishou-W at HKD 219 million, and SMIC at HKD 217 million [1] - Significant net sells included the Tracker Fund of Hong Kong at HKD 14.53 billion, Hang Seng China Enterprises at HKD 6.326 billion, and Southern Hang Seng Technology at HKD 1.453 billion [1] Group 2 - InnoCare Pharma announced a partnership with NVIDIA to promote the large-scale implementation of the 800 VDC power architecture in AI data centers, which offers significant advantages over traditional systems [3] - SMIC reported that Huaxia Fund recently reduced its holdings in SMIC's domestic stocks by 651,800 shares, representing a change of 0.03% [3] - Alibaba-W was appointed Zhang Yong, former chairman of Alibaba Group, to the China Business Advisory Committee of the Hong Kong Stock Exchange [3] - UBTECH Robotics won a robot procurement project worth over RMB 90.51 million from Miyi Automotive Technology, marking the largest humanoid robot order globally [3] - UBTECH also launched the full-size industrial humanoid robot Walker S2, featuring innovative hardware design and autonomous battery swapping technology for continuous operation [3]

Group 1 - The core viewpoint of the news is that InnoCare (2577.HK) has experienced significant stock price increases, rising nearly 31% last Friday and further increasing by over 27% today, reaching a historical high of 73.55 HKD with a trading volume of nearly 3 billion HKD [1][3] Group 2 - The company has recently announced a partnership with NVIDIA to jointly promote the large-scale implementation of the 800 VDC (800-volt direct current) power architecture in AI data centers [3] - This new power system, specifically designed by NVIDIA for future efficient power supply in megawatt-level computing infrastructure, offers significant advantages in system efficiency, thermal loss, and reliability compared to traditional 54V power systems [3] - The 800 VDC architecture is capable of supporting a 100 to 1000 times increase in AI computing power [3]

Core Viewpoint - InnoCare (2577.HK) has experienced significant stock price increases, rising nearly 31% last Friday and further expanding by over 24% today, reaching HKD 71.42, with a trading volume of HKD 2.4 billion [1] Company Summary - InnoCare has recently announced a partnership with NVIDIA to jointly promote the large-scale implementation of the 800 VDC (800-volt direct current) power architecture in AI data centers [1] - The 800 VDC architecture is specifically designed by NVIDIA for future efficient power supply megawatt-level computing infrastructure, offering significant advantages in system efficiency, thermal loss, and reliability compared to traditional 54V power systems [1] - This new power system can support an enhancement of AI computing power by 100 to 1000 times [1]

Core Viewpoint - InnoCare (02577.HK) has seen a significant increase in stock price due to its collaboration with NVIDIA to promote the large-scale implementation of 800 VDC (800-volt direct current) power architecture in AI data centers [1] Company Performance - InnoCare's stock rose by 7.47% [1] - Huahong Semiconductor (01347.HK) experienced a 7.22% increase in stock price [1] - Shanghai Fudan (01385.HK) saw a 5.41% rise in its stock price [1]

格隆汇8月4日｜继上周五收涨近31%后，英诺赛科(2577.HK)今早再度高开4.17%，报60港元。消息面 上，公司宣布，已于近日与英伟达达成合作，联合推动800 VDC(800伏直流)电源架构在AI数据中心的 规模化落地。该架构是英伟达针对未来高效供电兆瓦级计算基础设施而专门设计的新一代电源系统，相 比传统54V电源，在系统效率、热损耗和可靠性方面具有显著优势，可支持AI算力100-1000倍的提升。 ...

Major Events - InnoCare (02577) collaborates with NVIDIA to promote the large-scale implementation of 800 VDC (800-volt direct current) power architecture in AI data centers [1] - Zijin Mining (02899) receives parliamentary approval for the mining lease of Akyem gold mine in Ghana [1] - Codi-B (02487) receives approval from the Hong Kong Department of Health for the market launch of CU-40102 (topical finasteride spray) [1] - China Ruyi (00136) plans to issue a total of 1.3 billion subscription shares at a discount of approximately 4.46%, raising about HKD 38.95 billion [1] - Sinopec Engineering (02386) signs a front-end engineering design (FEED) contract for a large green hydrogen project in Yanbu, Saudi Arabia [1] - China Resources Land (01109) and a joint venture with Shanghai Nanfang acquire the Shanghai Yaohua Road project and Yuqingli project for approximately HKD 24.47 billion [1] - Global Industrial Technology (01026) intends to acquire a shopping mall property in Guangzhou's Yuexiu District for no more than HKD 260 million [1] - Corning Jereh Pharmaceutical-B (09966) has its IND application for JSKN022 officially accepted by CDE [1] - Sea Harvest International (01308) signs a shipbuilding contract with Huanghai Shipbuilding for the construction of four container ships [1] - Yika (09923) successfully completes registration as a "credit card number and contract conclusion business operator" in Japan [1] Operating Performance - Far East Horizon (03360) reports a mid-term profit attributable to shareholders of HKD 2.164 billion, an increase of 3.77% year-on-year [1] - Xinyi Energy (03868) reports a mid-term profit attributable to equity holders of HKD 450 million, an increase of 23.43% year-on-year [1] - Xinyi Solar (00968) reports a mid-term profit attributable to shareholders of HKD 746 million, a decrease of 58.8% year-on-year [1] - Xinyi Glass (00868) reports a mid-term profit attributable to shareholders of HKD 1.013 billion, a decrease of 59.64% year-on-year [1] - Xiehe New Energy (00182) continues to reduce financing costs and improve financing capabilities, with significant growth in green electricity business in the first half of the year [1] - PCCW (00008) reports a mid-term loss attributable to shareholders of HKD 445 million, a year-on-year narrowing of 3.68% [1] - Huishang Bank (03698) issues a profit warning, expecting mid-term net profit of approximately HKD 9.3 billion, a year-on-year increase of about 3.8% [1] - BYD (01211) reports sales of approximately 2.4903 million new energy vehicles in the first seven months, a year-on-year increase of 27.35% [1] - NIO-SW (09866) reports cumulative vehicle deliveries of 806,700 units as of the end of July [1] - XPeng Motors-W (09868) delivers 36,717 smart electric vehicles in July, a year-on-year increase of 229% [1] - Li Auto-W (02015) delivers 30,731 new vehicles in July [1] - Geely Automobile (00175) reports total vehicle sales of 237,700 units in July, a year-on-year increase of approximately 58% [1] - Great Wall Motors (02333) reports total vehicle sales of approximately 674,200 units in the first seven months, a year-on-year increase of 3.57% [1]

Core Viewpoint - The collaboration between InnoCare (02577.HK) and NVIDIA aims to promote the large-scale implementation of the 800 VDC power architecture in AI data centers, which is designed for high-efficiency power supply in megawatt-level computing infrastructure [1] Group 1: Partnership and Technology - InnoCare has recently partnered with NVIDIA, a global leader in AI technology, to advance the 800 VDC power architecture [1] - The 800 VDC architecture is specifically designed by NVIDIA for future high-efficiency power supply systems, offering significant advantages over traditional 54V systems in terms of efficiency, thermal loss, and reliability [1] - This new power system can support an increase in AI computing power by 100 to 1000 times [1] Group 2: Product Features - InnoCare's third-generation GaN devices exhibit excellent characteristics such as high frequency, high efficiency, and high power density [1] - These GaN devices provide a complete power solution from 800V input to GPU terminals, covering a voltage range from 15V to 1200V [1] Group 3: Future Outlook - The integration of GaN technology with NVIDIA's 800 VDC power architecture is expected to facilitate a leap from kilowatt-level to megawatt-level AI data centers in the coming years [1] - This advancement is set to usher in a more efficient, reliable, and environmentally friendly era of AI computing [1]