人民财讯12月16日电，企查查APP显示，近日，深圳市云集智体科技有限公司成立，注册资本1000万 元，经营范围包含：人工智能应用软件开发；人工智能硬件销售；数字技术服务等。企查查股权穿透显 示，该公司由云从科技全资子公司广州云从长嬴投资有限公司等共同持股。 转自：证券时报 ...

人民财讯12月16日电，12月16日，云从科技(688327)与深业资本、深圳科技园联合宣布发起设立"深业 云从人工智能产业股权投资基金"，赣深母基金、宁波奇精集团等多家机构共同出资，首期规模3亿元。 基金将聚焦智能算力与可信数据空间等AI基础设施、城市治理/金融/制造等行业智能体以及具身智能生 态，依托深业集团、赣深产业基金、宁波奇精集团等出资方产业资源，推进AI技术在实体经济中的体 系化落地。 ...

Group 1: AI Industry Trends - The AI industry is transitioning from cloud-based computing to edge devices, with significant advancements in hardware capabilities being essential for practical applications [2][3] - High costs associated with AI inference and training are major barriers to widespread adoption, prompting a need for cost-effective solutions [3][4] - The demand for printed circuit boards (PCBs) is surging due to the AI boom, with leading companies like Pengding Holdings experiencing increased profitability and market valuation [4][5] Group 2: Company Innovations and Strategies - AMD has introduced solutions that reduce the cost of deploying AI models on edge devices, making advanced AI applications more accessible across various sectors [2] - Midea Group is focusing on the integration of AI and robotics, with a strong emphasis on developing humanoid robots for industrial and commercial applications [5][6] - Companies like Haalo are exploring the Robotaxi market, indicating a growing recognition of the commercial viability of AI-driven transportation solutions [9][10] Group 3: Challenges and Opportunities - The AI sector faces challenges in commercializing applications and controlling costs, with many companies yet to achieve profitability [8][9] - The integration of AI into traditional industries is seen as a strategic necessity for survival and growth, with companies like Liaoning Chengda pivoting towards technology-driven sectors [12] - The need for high-quality data and deep industry understanding is critical for companies to maintain a competitive edge in the evolving AI landscape [11][12]

近日，云从科技(688327)联合上海交通大学、香港科技大学(广州)等多所高校及科研机构，共同发布全 球首个针对"空对空"场景的百万级多模态反无人机视觉追踪基准UAV-Anti-UAV，同步推出基于Mamba 架构的强力基线模型MambaSTS。 针对视频追踪的因果特性，模型还优化为单向扫描机制，更契合实时追踪需求。在50个主流深度追踪算 法的评测中，MambaSTS表现突出，在全部5项核心指标上均排名第一，AUC达0.437、mACC达0.443， 较第二名领先6.6个百分点。 值得关注的是，该模型在传统"空对地""地对空"数据集上同样取得SOTA性能，证明其具备强大的泛化 能力，并非局限于特定场景的"过拟合"模型。 在业内看来，UAV-Anti-UAV基准的发布，为行业提供了统一的技术评测标准，将加速反无人机技术的 迭代升级；MambaSTS模型的技术突破，则为实战化应用提供了高效解决方案，可广泛应用于安防监 控、空域管理、重大活动安保等场景。 据介绍，该成果填补了低空动态追踪领域的技术空白，为反无人机实战化应用提供了核心支撑，标志着 我国在多模态大模型与低空安全技术融合领域取得重要突破。 随着低空经济的 ...

Group 1 - The AI PC concept sector experienced a decline of 1.07%, ranking among the top losers in the concept sector, with notable declines from companies such as Shenghong Technology and Jiangbolong [1] - Among the AI PC concept stocks, four companies saw price increases, with Aoshikang, Siquan New Materials, and Yuncong Technology rising by 3.95%, 3.42%, and 1.00% respectively [1] - The top-performing concept sectors today included Hainan Free Trade Zone, which rose by 6.72%, and Duty-Free Shops, which increased by 3.01% [1] Group 2 - The AI PC concept sector faced a net outflow of 3.146 billion yuan, with 42 stocks experiencing net outflows, and six stocks seeing outflows exceeding 100 million yuan [1] - Shenghong Technology led the net outflow with 1.084 billion yuan, followed by Demingli and Changying Precision with net outflows of 324.21 million yuan and 217.60 million yuan respectively [1] - The stocks with the highest net inflows included Zhongshi Technology, with an inflow of 16.21 million yuan, and Shengli Precision, with an inflow of 9.20 million yuan [2]

Core Insights - The global AI agent market is experiencing explosive growth, with significant advancements in AI capabilities for mobile operations [1][3]. Group 1: AI Agent Development - Zhiyu announced the open-source of its core AI agent model AutoGLM, which can perform complex mobile operations such as food ordering and flight booking [2]. - AutoGLM supports over 50 high-frequency Chinese applications, marking it as the first AI agent with mobile operation capabilities [2]. - Zhiyu's GLM-4.5 ranked third globally in a comprehensive evaluation of AI capabilities, indicating strong market recognition [2]. Group 2: Market Growth Projections - The global AI agent market is projected to reach $11.3 billion by 2025 and nearly $50 billion by 2030, with a compound annual growth rate (CAGR) exceeding 30% in China from 2025 to 2030 [3][6]. - The Chinese AI agent market is expected to grow from 6.9 billion yuan in 2025 to nearly 30 billion yuan by 2030 [3]. Group 3: Capital Expenditure Trends - Major global tech companies are increasing capital expenditures to accelerate the commercial application of generative AI [5]. - China's four major cloud service providers have spent approximately $45 billion on capital expenditures over the past 12 months [6]. - The capital expenditure for AI agent concept stocks in the A-share market is projected to reach 26.53 billion yuan in 2024, a year-on-year increase of over 10% [6]. Group 4: Stock Performance and Institutional Interest - The average annual increase for AI agent concept stocks in the A-share market exceeds 25%, with over half of the stocks rising more than 10% [7]. - Eight concept stocks have seen significant institutional interest, with Shanghai Steel Union receiving over 150 institutional surveys this year [8]. - Financing balances for AI agent concept stocks have increased by nearly 45% compared to the end of last year, indicating strong investor interest [7].

Core Insights - The company aims to build an "AI infrastructure" that serves as a stable and powerful computing foundation for its self-developed "Congrong large model" and external clients [1] - The "AI intelligent agent" is designed based on a "perception-cognition-decision" technology loop, enabling multi-modal interaction and intelligent business flow orchestration [1] - The combination of these elements creates a complete value chain: "computing power supports the model, the model drives the intelligent agent, and the intelligent agent empowers various industries" [1] - The company has achieved commercial implementation in multiple sectors, including finance, government, and manufacturing [1]

Core Insights - The global AI agent market is experiencing explosive growth, with projections indicating a market size of $11.3 billion by 2025 and nearly $50 billion by 2030 [4] - The introduction of AutoGLM by Zhiyu, an AI agent capable of mobile operations, is a significant development, allowing for complex tasks like food ordering and flight booking [2] - Major technology companies are increasing capital expenditures to accelerate the commercial application of generative AI, with Chinese cloud service providers spending approximately $45 billion over the past year [6] Market Growth - The AI agent market is expected to see a compound annual growth rate (CAGR) exceeding 30% from 2025 to 2030 in China, with the market size reaching approximately 30 billion yuan by 2030 [4] - The Chinese government's policy aims for over 70% adoption of new intelligent terminals and agents by 2027, and over 90% by 2030, positioning the intelligent economy as a key growth driver [4] Capital Expenditure Trends - Capital expenditures in the AI sector are on the rise, with a projected total of 26.53 billion yuan for 2024, reflecting a year-on-year increase of over 10% [6] - The A-share market shows a significant increase in capital expenditure among AI agent concept stocks, with a total of nearly 30.5 billion yuan spent in the first three quarters of 2025, up nearly 15% from the previous year [6] Stock Performance - The average annual increase for AI agent concept stocks in the A-share market exceeds 25%, with over half of the stocks rising more than 10% this year [7] - Eight specific stocks have seen significant institutional interest, with Shanghai Steel Union receiving over 150 institutional surveys this year [8][9] - Financing balances for AI agent concept stocks have increased by nearly 45% compared to the end of last year, indicating strong investor interest [7]

证券日报网讯12月8日，云从科技在互动平台回答投资者提问时表示，公司在坚持主业稳健发展的同 时，始终关注行业前沿动态，秉持开放态度，根据战略规划及业务需求，评估各类有利于提升公司核心 竞争力和长期价值的机会。若有相关重大事项和进展，公司将严格按照有关法律法规的规定及时履行信 息披露义务。 ...

Core Insights - Yuncong Technology has officially received the "Preferred Partner" certification from Huawei Ascend, marking a new phase of deep integration between the two companies and indicating a shift towards collaborative efforts among domestic software and hardware manufacturers to build a self-controlled intelligent computing industry [1][2] Group 1: Partnership Development - The collaboration between Yuncong Technology and Huawei Ascend began in 2019, evolving from initial technical adaptation to a strategic symbiosis [2] - The partnership aims to leverage Huawei's computing power advantages and Yuncong's agile capabilities in AI native application development [1][2] Group 2: Technological Advancements - Yuncong Technology's self-developed "Congrong" large model has achieved bottom-layer adaptation and full-stack optimization with Huawei's Ascend Atlas series hardware, effectively breaking through performance bottlenecks in heterogeneous computing [1] - The "soft and hard integration" collaboration is accelerating the transition of AI from the laboratory to production lines, focusing on commercializing AI native applications in sectors such as intelligent manufacturing, smart finance, and urban governance [2] Group 3: Strategic Goals - The partnership aims to create a closed loop between computing power, algorithms, and application scenarios, ensuring that domestic AI applications are not only functional but also user-friendly, addressing industry pain points [2] - Yuncong Technology views becoming a Huawei Ascend preferred partner as a result of mutual strategic trust, with plans to define new standards for the AI industry by combining strengths in computing infrastructure and decision-making technologies [2]