Investment Rating - The report maintains a positive outlook on the ASIC industry, indicating a favorable investment rating for the sector [2]. Core Insights - The report highlights the significant cost-effectiveness and efficiency advantages of ASICs over GPUs, particularly in the context of AI model inference, with Google's TPU v5 demonstrating an energy efficiency ratio 1.46 times that of NVIDIA's H200 [3][19]. - The increasing penetration of AI applications is driving a surge in inference demand, expanding the market for ASICs, with projections indicating the global AI ASIC market could reach $125 billion by 2028 [3][32]. - The report emphasizes the complexity of ASIC design, underscoring the critical role of design service providers like Broadcom and Marvell, which are expected to benefit from the growing demand for custom ASIC solutions [4][44]. Summary by Sections 1. Demand Driven by Large Model Inference - The global AI chip market is projected to reach $500 billion by 2028-2030, with significant growth in AI infrastructure spending anticipated [13]. - ASICs are specialized chips that offer strong cost and efficiency advantages, particularly in specific applications like text and video inference [14][19]. - The report notes that the demand for ASICs is expected to rise sharply due to the increasing consumption of tokens in AI applications, exemplified by the rapid growth of ChatGPT's user engagement [25][31]. 2. High Complexity of ASIC Design and Value of Service Providers - ASIC design involves a complex supply chain, with cloud vendors often relying on specialized design service providers for chip architecture and optimization [41][44]. - Broadcom's ASIC revenue is projected to exceed $12 billion in 2024, driven by the success of its TPU designs for Google and other clients [60]. - The report identifies the importance of a complete IP system and design experience as key factors for service providers to secure new orders in the ASIC market [63]. 3. Domestic Developments: Not Just Following Trends - Leading Chinese cloud providers like Alibaba and Baidu are making significant strides in self-developed ASICs, indicating a robust domestic ecosystem [3][4]. - The report highlights the emergence of domestic design service providers such as Chipone and Aowei Technology, which are positioned to capitalize on the growing demand for ASICs [3][4]. - The trends of PD separation and supernodes are identified as critical developments in the domestic ASIC landscape, with companies like Huawei and Haiguang leading the way [4][44]. 4. Key Trends in Domestic ASIC Development - PD separation involves using different chips for prefill and decode tasks, enhancing efficiency in specific applications [4]. - Supernodes are being developed to create unified computing systems through high-bandwidth interconnections, with early implementations seen in domestic companies [4][44].

Group 1 - Huawei has launched new super nodes and multiple Ascend series chips, with the Ascend 950-based super node expected to be the strongest globally, surpassing NVIDIA's NVL576 system set to release in 2027 [1] - The Ascend 960-based super node is also scheduled for release in Q4 2027, promising substantial computing power [1] - Huawei introduced its self-developed low-cost HBM, supporting more precision formats like FP8, and announced the "Lingqu" interconnection protocol for super nodes, capable of forming clusters of over 500,000 cards [1] Group 2 - The first quantum computing company in China, Benyuan Quantum, has initiated an IPO, focusing on building self-controlled engineering quantum computers with a full-stack layout covering quantum chips and quantum computing measurement and control integrated machines [1] - NVIDIA plans to invest $5 billion in Intel, while Microsoft is investing $7.3 billion to create the "world's strongest AI data center," indicating a sustained high demand for global computing power investments, with AI applications expected to accelerate benefits [1] Group 3 - The software ETF (515230) tracks the software index (H30202), which selects listed companies involved in system software, application software development, and related services to reflect the overall performance of the software industry [1] - The index focuses on companies with outstanding technological innovation capabilities and high market share, characterized by significant growth and volatility [1] - Investors without stock accounts can consider the Guotai CSI Software ETF Connect A (012636) and Guotai CSI Software ETF Connect C (012637) [1]

Core Viewpoint - The semiconductor industry is experiencing positive developments driven by advancements in AI technology and domestic initiatives to enhance self-sufficiency in semiconductor manufacturing [3][16][18]. Group 1: Industry Developments - Huawei has unveiled its AI computing power landscape, introducing a series of chips including Ascend 950, Yitian 960, and Yitian 970 [3]. - Alibaba is actively advancing a 380 billion yuan investment in AI infrastructure [3]. - Progress has been made in testing domestic photolithography machines, indicating advancements in local semiconductor capabilities [3]. Group 2: Semiconductor Industry Chain - The semiconductor industry chain is divided into three main segments: upstream (basic materials), midstream (manufacturing), and downstream (applications) [5][6]. - Upstream segments include chip design, semiconductor equipment, and materials, which are crucial for chip manufacturing and packaging [7]. Group 3: Demand Cycles - The semiconductor demand cycle is influenced by global economic fluctuations, typically spanning 2 to 3 years [9]. - The innovation cycle, driven by technological advancements such as AI, generally requires 5 to 10 years for significant upgrades in chip performance [10]. - The domestic substitution cycle, aimed at increasing self-sufficiency in semiconductor technology, is projected to take 10 to 20 years [11]. Group 4: Market Growth - The global semiconductor market reached a size of 346 billion dollars in the first half of the year, reflecting an 18.9% year-on-year growth [13]. - The demand for semiconductors is expected to continue growing, particularly due to the AI boom, with the Chinese AI chip market projected to reach 153 billion yuan by 2025 [16]. Group 5: Domestic Substitution - There is significant potential for domestic substitution in the semiconductor sector, especially in equipment and materials, as the gap between domestic and foreign technology remains substantial [18]. - Supportive policies and funding initiatives are being implemented to accelerate the development of a self-sufficient semiconductor ecosystem in China [18].

Group 1 - The core viewpoint of the article highlights the resurgence of domestic computing power, with the domestic chip ETF, Guotai (589100), rising over 5% during trading [1][3] - The State Administration for Market Regulation announced an investigation into NVIDIA for violating China's antitrust laws, while Huawei launched new supernodes and Ascend series chips, indicating a strong domestic push for AI computing power [3] - The development of domestic computing power is supported by continuous iterations of domestic chips like Huawei's Ascend, alongside increasing capital expenditures from major domestic internet companies, indicating a robust demand [3][4] Group 2 - The article suggests that a new arms race in AI is underway, with a shift in demand from AI training to inference, indicating stronger and more sustained investment compared to previous cycles [4] - The improving market risk appetite suggests that there is still room for valuation increases within the relevant industry chain, encouraging investors to pay attention to the Guotai chip ETF (589100) and the chip ETF (512760) [4]

Core Viewpoint - Huawei is positioning itself as a leader in AI infrastructure by introducing advanced computing capabilities and innovative AI models, aiming to simplify complex processes for enterprises while enhancing their operational efficiency [5][6][26]. Group 1: AI Infrastructure and Innovations - Huawei announced a roadmap for multiple chip releases and supernode advancements over the next three years, aiming to create the "world's strongest supernode" in AI computing [5]. - The CloudMatrix supernode specifications will upgrade from 384 cards to 8192 cards, enabling the formation of super-large clusters of 500,000 to 1,000,000 cards, significantly enhancing AI computing power [7][8]. - The CloudMatrix384 can support 384 Ascend NPUs and 192 Kunpeng CPUs, facilitating the training of large models and improving inference performance by pooling resources [7][8]. Group 2: Strategic Focus and Market Position - Huawei Cloud's strategy emphasizes "system-level innovation" and a focus on various industries, which is seen as a proactive response to global AI competition [6][7]. - The company has achieved a 268% increase in AI computing scale compared to the previous year, with the number of Ascend AI cloud customers rising from 321 to 1805 [26]. Group 3: Industry Applications and Case Studies - Huawei Cloud has successfully implemented AI solutions in various sectors, such as transportation and manufacturing, demonstrating significant improvements in operational efficiency and predictive maintenance [12][24][25]. - The integration of AI models like Pangu has led to enhanced accuracy in traffic prediction and operational processes, showcasing the practical benefits of AI in real-world applications [12][24]. Group 4: Global Reach and Data Solutions - Huawei Cloud operates in 34 geographical regions with 101 availability zones, providing a global network that enhances data processing and AI application development [20][21]. - The company has improved data integration efficiency for clients like Neogrid, enabling faster decision-making through real-time data access [22]. Group 5: Future Vision and Commitment - Huawei emphasizes the importance of collaboration across the AI industry to build a future-oriented ecosystem that benefits all stakeholders [26]. - The company's commitment to simplifying complex processes for clients while managing intricate data and AI systems reflects its long-term vision for AI and digital transformation [17][26].

更多集体接待日详情，请点击：https://rs.p5w.net/html/175611728073329.shtml 恒为科技在2025年上海辖区上市公司集体接待日暨中报业绩说明会活动上回答投资者提问称，公司与华 为昇腾的生态合作主要体现在：基于昇腾算力集群的运营，客户模型优化与训练；以及基于昇腾的推算 算力一体机等方面。昇腾950是华为近期新发布的产品规划，目前还未向市场投放，我们关注华为昇腾 新一代架构GPU的发布与应用，期待能进一步与相关AI芯片的系统架构上展开更深入合作。 ...

Core Viewpoint - Huawei's Full Connect 2025 Conference showcased the Lingqu Unified Interconnection Protocol, announcing the launch of Ascend 950/960/970 and Kunpeng 950/960 over the next three years, highlighting the gradual enhancement of domestic AI computing chip capabilities amid US-China tensions [9][58]. Group 1: AI Computing Chip Development - The Ascend NPU roadmap includes the release of Ascend 950 (PR and DT versions) in 2026, followed by 960 in 2027 and 970 in 2028, with significant performance improvements [15][20]. - The Kunpeng CPU will see the launch of Kunpeng 950 in late 2026 and Kunpeng 960 in early 2028, supporting advanced computing needs [20][34]. - Domestic chip manufacturers like Haiguang and Cambricon are projecting substantial revenue growth, with Haiguang targeting a CAGR of 44% over three years [3][58]. Group 2: Advanced Manufacturing and Semiconductor Industry - The domestic lithography machine industry is focusing on complete machines and related components, with expectations for advanced process expansion by 2026 [3][59]. - The domestic semiconductor industry is expected to benefit from the acceleration of independent and controllable demands, particularly in advanced logic and storage production lines [3][62]. Group 3: Storage and Edge Computing - The demand for inference and edge computing storage is increasing, with significant growth expected in AI PCs, smartphones, and wearable devices by 2026 [4][58]. - Domestic manufacturers are enhancing their enterprise storage product lines, with companies like Jiangbolong and Baiwei Storage launching new enterprise-level solutions [4][58]. Group 4: Investment Recommendations - Investment opportunities are suggested in AI computing chips, high-end chip manufacturing, packaging, storage, and related equipment and materials [5].