Core Insights - Deepexi Technology has officially listed on the Hong Kong Stock Exchange, becoming the first "enterprise-level large model AI application" stock, with an oversubscription rate of 7569.83 times, setting a record for the Hong Kong main board [1][4] - The company focuses on enterprise-level AI solutions, aiming to enhance productivity by integrating AI decision-making systems with proprietary enterprise data [1][4] - Despite high initial demand and a significant market valuation, the company faces challenges in achieving positive cash flow due to high R&D and personnel expenses [1][10] Group 1: Market Position and Competition - The AI market is witnessing intensified competition, with companies like Fourth Paradigm, Yitu Technology, and SenseTime accelerating their enterprise AI strategies [1] - Deepexi's business model is compared to Palantir, highlighting the shift from AI computing power to AI application, indicating a broader trend in the industry [2][13] - The enterprise AI market in China is projected to reach 239.4 billion RMB by 2029, with a compound annual growth rate of 44%, positioning Deepexi as a key player in this evolving landscape [14] Group 2: Financial Performance and Growth - Deepexi's revenue has shown significant growth, increasing from 100 million RMB in 2022 to 132 million RMB in the first half of 2025, with a year-on-year growth rate of 118% [9][10] - The company's gross margin improved from 29.4% in 2022 to 55% in the first half of 2025, driven by the high value of its FastAGI solutions [7][9] - Despite revenue growth, Deepexi reported net losses of 655 million RMB in 2022, 503 million RMB in 2023, and 1.255 billion RMB in 2024, indicating ongoing high investment in R&D [10] Group 3: Technological Foundation and Product Offerings - Deepexi has developed two main technological platforms: FastData Foil for data integration and Deepexi's enterprise-level model platform for customized AI solutions [5] - The company offers two primary product lines: FastData for data governance and FastAGI for AI solutions across various operational scenarios [5][8] - As of mid-2025, FastAGI accounted for 55.3% of the company's revenue, highlighting its role as a core growth driver [7] Group 4: Leadership and Investment - The founding team of Deepexi has a strong background, with CEO Zhao Jiehui previously serving as a core technology expert at Huawei and general manager at Alibaba Cloud [11][12] - The company has attracted significant investment from prominent firms, including Hillhouse Capital and IDG Capital, and has completed 10 rounds of financing prior to its IPO [12] - The net proceeds from the IPO will be allocated to enhancing R&D capabilities, expanding the sales network, and exploring overseas business opportunities [12]

Core Viewpoint - Dipu Technology, a Beijing-based enterprise AI solution provider, has successfully listed on the Hong Kong Stock Exchange, raising approximately HKD 609.8 million (around RMB 558 million) with a subscription rate of 7569.83 times, setting a record for oversubscription in the history of the Hong Kong main board [1][3]. Company Overview - Founded in 2018, Dipu Technology was established by Zhao Jiehui, a former core technology expert at Huawei, and Yang Lei, who held senior positions at Huawei [3]. - The company initially focused on "data middle platform" products, offering FastData enterprise-level data intelligence solutions [3]. - In late 2023, Dipu launched the FastAGI enterprise-level AI solution, which is expected to contribute over 55.3% of its revenue by mid-2025 [3][4]. Market Position - The enterprise-level AI application solutions market is vast, with Dipu holding a market share of 0.6% in 2024 and ranking fifth in the enterprise-level large model AI application segment with a 4.2% market share [4][8]. - The company aims to provide a one-stop enterprise-level large model solution, integrating model development, data platforms, and computing power optimization [8]. Financial Performance - Dipu's revenue from 2022 to mid-2025 shows a significant growth trajectory, with revenues of RMB 1.00 billion, RMB 1.29 billion, RMB 2.43 billion, and RMB 1.32 billion respectively, reflecting year-on-year growth rates of 28.4% and 88.3% for 2023 and 2024 [8][11]. - The company has reported cumulative losses exceeding RMB 2.7 billion over three and a half years, with a narrowing net loss of RMB 3.08 billion in the first half of 2025 [11][12]. Client Base - Dipu's clients span various sectors, including consumer retail, manufacturing, healthcare, and transportation, with notable clients such as China National Petroleum, China Petroleum & Chemical Corporation, and SAIC-GM-Wuling [5][18]. - The company has seen a decrease in client concentration, with the top five clients contributing 31.7% of total revenue in the first half of 2025 [18]. Research and Development - Dipu plans to allocate 40% of its raised funds to enhance R&D capabilities, 30% to expand its sales network, and 15% for overseas business expansion [7]. - The company has a significant focus on R&D, with expenditures representing a large portion of its revenue, although the size of its R&D team has decreased from 238 to 147 members [11][17]. Competitive Landscape - Dipu's business model and client types are similar to those of Hong Kong-listed Fourth Paradigm, which reported a revenue of RMB 5.261 billion in 2024, significantly higher than Dipu's [12][13]. - The competitive landscape in China's enterprise-level AI application market is fragmented, with over 200 participants and the top five companies holding more than 30% of the market share [8].

Core Insights - Neolix has completed a D-round financing of over $600 million, led by Stone Venture from the UAE, indicating growing interest from Middle Eastern capital in Chinese tech companies [1] - Since 2025, Middle Eastern investments in China have exceeded 10 cases, totaling over $4 billion, covering various sectors such as finance, healthcare, new energy, and hard technology [1][4] Group 1: Neolix's Market Position - Neolix is a leader in the autonomous vehicle market, having established a strong presence with over 300 delivery vehicles deployed across more than 300 cities globally [2] - The company has received the first license for unmanned delivery vehicles in the UAE, marking a significant milestone for its operations in the Middle East [2] Group 2: Strategic Collaborations - Neolix has formed a strategic partnership with K2 Group, a UAE state-owned AI technology company, to accelerate the application of autonomous driving technology in the region [3] - This collaboration aims to develop delivery solutions tailored to the unique climate and regulatory standards of the Middle East [3] Group 3: Investment Trends - Middle Eastern capital is increasingly focusing on structural cooperation and project-based investments rather than traditional financial investments, reflecting a shift in investment logic [1][6] - The trend shows a preference for investments that contribute to local economic diversification and infrastructure development [6][7] Group 4: Future Investment Characteristics - Future investments from Middle Eastern countries are expected to prioritize three key characteristics: industry certainty, internationalization capability, and controllable exit channels [8]

Group 1 - The Hong Kong stock market opened higher on October 24, with the Hang Seng Index rising over 1%, driven by a rebound in tech stocks and the pre-sale of Alibaba's first self-developed AI glasses [1][4] - Alibaba's Quark AI glasses, priced at 3,699 yuan for VIP members and 3,999 yuan for regular consumers, feature advanced capabilities such as navigation and secure payment systems [4][5] - The Hong Kong Internet ETF (513770) saw a strong opening, with a price increase of 0.84% and a significant net inflow of 217 million yuan over the past two days [2][3] Group 2 - Major companies like Alibaba, Xiaomi, and Huawei are entering the AI glasses market, indicating a rapid evolution of the AI terminal ecosystem, with expectations for significant advancements by 2026 [4][5] - The Hong Kong Internet ETF tracks the CSI Hong Kong Internet Index, with Alibaba, Tencent, and Xiaomi being the top three holdings, accounting for 18.11%, 16.16%, and 11.06% of the index, respectively [4][5] - The internet sector in Hong Kong has shown higher elasticity this year, with the CSI Hong Kong Internet Index outperforming the Hang Seng Tech Index [5][6]

Core Insights - The "OpenMLDB: Key Technologies and System Platform for Enterprise-level Machine Learning Database" project, developed by 4Paradigm in collaboration with Shanghai Jiao Tong University and Tsinghua University, won the CCF Technology Achievement Award for its significant breakthroughs in enterprise-level machine learning data management [1][5]. Group 1: Technological Breakthroughs - The OpenMLDB project addresses significant challenges in data management for AI applications, particularly in core business systems, focusing on data consistency, real-time processing, and efficient preparation of large training datasets [2]. - Innovations include achieving data calculation consistency between online inference and offline training, enhancing the accuracy and reliability of AI applications [2]. - The project has developed millisecond-level real-time data processing technology, supporting immediate responses in critical areas such as financial risk control and intelligent recommendations [2]. - It has also improved the efficiency of batch preparation for large-scale training data, significantly shortening feature engineering and model iteration cycles [2][3]. Group 2: Economic Impact - The OpenMLDB project has been successfully deployed in various industries, including finance and e-commerce, demonstrating its capability to enhance system stability, data accuracy, and processing performance [4]. - In a leading bank, the project enabled millisecond-level real-time feature computation, effectively reducing financial losses from fraudulent transactions [4]. - The project supports the online recommendation system of Vipshop, improving recommendation accuracy and user experience, which directly contributes to an increase in GMV [4]. - The successful application of OpenMLDB has led to significant efficiency improvements and cost optimization for its users, providing a reusable technical pathway for intelligent transformation in related industries [4]. Group 3: Open Source and Innovation - 4Paradigm actively operates and builds the OpenMLDB open-source community, promoting the open and collaborative development of technology [5]. - This initiative accelerates the dissemination and popularization of project results, gathering global developer insights to advance enterprise-level machine learning data management technology [5]. - OpenMLDB has established itself as a domestically developed open-source solution with independent intellectual property rights, lowering the barriers for enterprises and developers to adopt advanced AI data management technologies [5]. - The CCF Technology Achievement Award recognizes significant discoveries and innovations in computer science and technology, with the OpenMLDB project standing out among 94 submissions [5].

Investment Rating - The industry investment rating is "Recommended," indicating an expected increase in the industry index exceeding the benchmark index by more than 5% in the next 3-6 months [16]. Core Insights - The report highlights the launch of DeepSeek-OCR, a visual text compression model that utilizes optical compression technology for efficient parsing and cross-language processing of multimodal documents. This model is expected to significantly enhance data processing capabilities in various fields [2][5]. - DeepSeek-OCR achieves a compression ratio of 10:1 while maintaining 97% near-lossless decoding accuracy, showcasing its efficiency and performance in handling large volumes of data [5]. - The model's architecture includes a visual encoder (DeepEncoder) and a lightweight decoder (DeepSeek3B-MoE), which together optimize the processing of visual tokens, reducing computational overhead [5]. Industry Overview - The computer industry comprises 337 listed companies with a total market capitalization of approximately 59,939.50 billion and a circulating market value of about 54,322.71 billion [2]. - The absolute performance of the industry over the past 12 months has been 31.4%, while the relative performance against the benchmark has been 15.0% [3]. - The report emphasizes the potential of DeepSeek-OCR in various application scenarios, including finance, education, legal, medical, e-commerce, security, industrial, and military sectors, indicating a broad market opportunity [5]. Investment Recommendations - The report suggests focusing on specific segments within the AI industry, including domestic computing power providers and enterprise services across various applications [5]. - Key companies to watch include Alibaba, Cambricon, and Kingsoft Office, among others, across different sectors such as finance, education, and healthcare [5].

Investment Rating - Investment advice is optimistic about the long-term growth of domestic computing power and AI applications, particularly focusing on key targets in AI video and image-related sectors [3][7]. Core Insights - OpenAI's annual developer conference showcased the development of ChatGPT as an AI operating system, with significant user engagement, including 800 million weekly active users and over 4 million developers utilizing OpenAI [3][8]. - The introduction of new products such as the AI model GPT-5 Pro and the AgentKi platform for building AI agents indicates a strong push towards enhancing AI capabilities [3][8]. - The Apps SDK allows developers to create interactive applications within ChatGPT, emphasizing context memory and understanding, which positions ChatGPT as a central entry point for users [9][10]. Summary by Sections - **Investment Highlights**: The report highlights the potential of domestic computing power and AI applications, with key targets including companies like ArcSoft, Hangzhou Arcvideo Technology, and others in the AI video and image sector [3][7]. - **User Engagement**: ChatGPT has reached 800 million weekly active users, with increasing adoption across various sectors, including consumers, developers, enterprises, and governments [8]. - **Product Development**: OpenAI announced several new products, including GPT-5 Pro and AgentKi, aimed at enhancing the functionality and usability of AI applications [8][10]. - **Developer Tools**: The Apps SDK and AgentKit provide developers with tools to create applications and agents, streamlining the development process and improving integration with existing systems [9][10].

Investment Rating - The report maintains a positive outlook on the AI video generation sector, particularly highlighting the potential of the Sora App and Sora2 model from OpenAI, which is seen as a transformative moment in the industry [4][8]. Core Insights - OpenAI's launch of the Sora App and Sora2 model on September 30 marks a significant shift in AI technology from text and images to high-quality video generation and social applications, creating a new ecosystem of "AI creation + social interaction" [4][9]. - The Sora App quickly gained popularity, reaching third place in the local free app rankings in the US, indicating strong market interest and potential for growth [4][9]. - The Sora2 model has achieved breakthroughs in physical motion, character performance, and consistency, allowing for realistic simulations of complex movements and coherent multi-shot scenes [10][11]. Summary by Sections Investment Highlights - The report emphasizes the long-term growth potential of domestic computing power and AI applications, particularly in the context of increasing demand for computing resources from cloud service providers and data center operators [4][8]. - Key investment targets identified include companies involved in AI video and image technology, such as ArcSoft Corporation, Hangzhou Arcvideo Technology, and others [4][8]. Sora App and Sora2 Model - The Sora App is designed around "pure AI content + strong social" and features innovative functions like "Cameo" for creating digital avatars and "Remix" for easy re-creation of content, which encourages user creativity [11]. - The current limitations of the Sora2 model include fixed video length of 10 seconds, resolution of 360P, and challenges in generating complex scenes accurately [10][11]. Market Position and Competitive Landscape - The report suggests that the Sora App and Sora2 model could disrupt traditional short video platforms by shifting the content creation paradigm from consumption to active user participation [9][11]. - The report also notes the potential for rapid social network growth through a viral mechanism that allows users to share invitations to join the platform [11].

Investment Rating - The report suggests a cautious subscription for the company, 聚水潭 (6687.HK) [1][3] Core Insights - 聚水潭, established in 2014, has become the largest e-commerce SaaS and ERP provider in China, holding a market share of 24.4% in the e-commerce SaaS market and 8.7% in the overall SaaS market by 2024 [2][3] - The Chinese e-commerce market has rapidly evolved, with online retail sales expected to grow from RMB 11.8 trillion in 2020 to RMB 16.5 trillion in 2024, reflecting a compound annual growth rate (CAGR) of 8.9% [2] - The company's total revenue and SaaS product turnover reached RMB 9.1 billion and RMB 13.0 billion respectively in 2024, with CAGRs of 31.9% and 32.6% from 2022 to 2024 [3] - The company transitioned from a net loss of RMB 4.9 billion in 2024 to a net profit of RMB 10.58 million in the same year, improving its net profit margin from a loss rate of 70.3% in 2023 to a profit margin of 1.2% in 2024 [3] - The estimated market capitalization post-IPO is approximately HKD 10.4 billion, with a price-to-sales (PS) ratio of about 10.0X based on 2024 revenue [3] Summary by Sections IPO Details - The IPO price is set at HKD 30.60, with a total fundraising amount of HKD 19.38 billion [1] - The total number of shares offered is 6.817 million, with 90% allocated for international placement and 10% for public offering [1] Company Overview - 聚水潭 has expanded its workforce from 9 employees at inception to over 2,500, establishing more than 100 offline service points across over 500 towns in China [2] - The company provides a comprehensive suite of SaaS products and services aimed at enhancing business capabilities and reducing operational costs for clients [3]

Group 1 - The article discusses the performance and outlook of the two-wheeled vehicle business of Ninebot Company, focusing on its core target audience and providing a full lifecycle user experience [3] - The company is expected to maintain a growth trajectory supported by internationalization and diversification strategies, with projected net profits of 2 billion, 2.7 billion, and 3.5 billion yuan for 2025-2027, corresponding to PE ratios of 23, 18, and 14 times [4] Group 2 - The report highlights the performance of public REITs in China, with a total of 75 products and a combined issuance scale of 196.62 billion yuan as of September 30, 2025, noting a downward trend in secondary market prices [10] - The article emphasizes the ongoing recovery of industrial profits in China, with a year-on-year growth rate improvement observed from January to August 2025, and suggests that domestic exports may remain resilient [12] - The report on the controllable nuclear fusion industry indicates positive progress with the successful installation of the BEST Dewar base, anticipating a period of intensive bidding and highlighting the long-term growth potential of the industry [14] Group 3 - The article outlines three new variables driving the recent rise in gold prices during the National Day holiday, including concerns over U.S. government shutdowns, political changes in Japan and France, and increased inflows into gold ETFs [22] - The credit bond market is experiencing growth, with a total outstanding balance of 30.49 trillion yuan and a net financing of 139.89 billion yuan in September 2025, indicating a widening credit spread [24] - The report indicates a sustained increase in market risk appetite, with over 60% of stocks in the CSI 300 index rising in the past month, reflecting optimistic market sentiment [27]