光伏资讯 | PV-info 近日，国家发展改革委、国家能源局下发关于推进"人工智能+"能源高质量发展的实施意见。 文件特别提出，人工智能+新能源典型应用场景。 针对新能源出力波动性与间歇性的问题，加快在高精度功率预测、电力市场、场站智慧运营、新能源规划、项目后评 价等方向的人工智能应用，持续推动新能源关键材料及产品不断迭代和创新，推动复杂场景及转折性天气下功率预测 大模型在更小尺度、更高精准度方向发展，支撑广域新能源资源协同优化，促进偏远地区新能源场站智能运维发展， 打造"气象预测+功率预测+智慧交易+智能运维"一体化新能源智能生产模式，全力支撑新能源稳定供给。 DeepSeek-R1具备多模态处理与动态注意力机制，可高效整合文本、代码、图像等多类型数据，龙源电力结合新能源 数字化平台构建垂直领域智能体，配合RAG（检索增强生成）架构，利用平台全量汇聚的生产业务数据、专业技术文 档，形成龙源电力专属知识库，为龙源电力新能源生产运营决策提供有力支持。 今年以来，以DeepSeek系列为首的人工智能大模型以惊人的速度渗透到互联网、能源、制造等核心产业领域。 能源央企积极响应国家人工智能战略，持续推进 "人工智能 ...

Market Overview - The A-share market experienced a significant drop on July 31, with the Shanghai Composite Index falling over 1% and closing below 3600 points, while the total trading volume approached 2 trillion yuan [1] - The Shanghai Composite Index closed at 3573.21 points, down 1.18%, the Shenzhen Component Index fell 1.73% to 11009.77 points, and the ChiNext Index decreased by 1.66% to 2328.31 points [1][2] Sector Performance - The overall market showed weakness, with nearly 4300 stocks declining, particularly in sectors such as coal, steel, oil, and non-ferrous metals [2] - The information security sector saw a strong performance, with stocks like Digital Certification and Nansheng Co. hitting the daily limit, while companies like Yonyou Network and Yongxin Zhicheng rose over 8% [2][4] - The assisted reproduction sector also surged, with stocks like Anke Bio and Gongtong Pharmaceutical reaching a 20% increase, and Guangshentang rising over 18% [11][12] AI Industry - The AI industry chain stocks were active, with companies like Fangsheng Co. and Dingtong Technology seeing significant gains, reflecting a growing interest in AI applications [7][8] - Domestic AI models have shown performance comparable to leading overseas models, with advancements in AI chip manufacturing by companies like Haiguang Information and Huawei [9] Policy Impact - The recent announcement of a child-rearing subsidy policy by the central government is expected to stimulate consumption in maternal and infant-related sectors, with an initial budget of approximately 90 billion yuan allocated for this purpose [13] - The implementation of this policy is anticipated to boost the birth rate and support industries related to assisted reproduction, genetic testing, and childcare services [13]

A股市场主要指数全线收跌。 A股今日（7月31日）大幅跳水，沪指跌超1%失守3600点，全A成交额逼近2万亿；港股亦下挫，恒生指数跌超1%。 具体来看，沪指盘中震荡下探，尾盘加速下行；深证成指、创业板指跌近2%。截至收盘，沪指跌1.18%报3573.21点，深证成指跌1.73%报11009.77点，创 业板指跌1.66%报2328.31点，上证50指数跌1.54%，沪深北三市合计成交19621亿元，较此前一日增加逾900亿元。 | 行情 | 资金净流入 | 涨跌分布 | | --- | --- | --- | | 上证指数 | 深证成指 | 北证50 | | 3573.21 | 11009.77 -42.51 -1.18% -193.26 -1.73% -16.77 -1.16% | 1422.87 | | 科创50 | 创业板指 | 万得全A | | 1047.87 | 2328.31 -10.70 -1.01% -39.37 -1.66% -76.50 -1.36% | 5567.10 | | 沪深300 | 中证500 | 中证A500 | | 4075.59 | 6226.34 -75.65 -1.8 ...

Market Overview - The A-share market experienced a significant decline, with the Shanghai Composite Index dropping over 1% and closing at 3573.21 points, while the Shenzhen Component Index fell 1.73% to 11009.77 points [1][2] - The total trading volume in the A-share market approached 2 trillion yuan, increasing by over 900 billion yuan compared to the previous day [1] Sector Performance - Resource sectors such as coal, steel, oil, and non-ferrous metals saw collective pullbacks, while brokerage, insurance, liquor, and banking sectors also declined [2] - Conversely, the information security sector surged, with stocks like Digital Certification and Nanxing Co. hitting the daily limit, and several others seeing significant gains [2][3] Information Security Sector - The information security concept saw strong performance, with *ST Guandao and Digital Certification both hitting the daily limit of 30% and 20% respectively, alongside notable gains from other companies in the sector [3][4] AI Industry - AI-related stocks were active, with companies like Fangsheng Co. and Dingtong Technology seeing substantial increases, reflecting a growing interest in AI applications [6][7] - The domestic AI industry is reportedly advancing, with local models achieving performance comparable to leading overseas models, and significant progress in AI chip development by companies like Huawei and Haiguang Information [8] Assisted Reproductive Technology - The assisted reproductive sector experienced a notable surge, with stocks like Anke Bio and Gongtong Pharmaceutical reaching daily limits of 20% [9][10] - The recent implementation of a childcare subsidy policy is expected to stimulate consumption in maternal and infant-related industries, including assisted reproduction and genetic testing [11][12]

Market Overview - A-shares experienced a significant drop on July 31, with the Shanghai Composite Index falling over 1% and closing at 3573.21 points, while the Shenzhen Component Index and ChiNext Index declined nearly 2% [1][2] - The total trading volume in the Shanghai and Shenzhen markets reached approximately 1.96 trillion yuan, an increase of over 90 billion yuan compared to the previous day [1][2] Sector Performance - The overall market showed weakness, with nearly 4300 stocks declining. Key sectors such as coal, steel, oil, and non-ferrous metals saw collective pullbacks, while brokerage, insurance, liquor, and banking sectors also fell [2] - Conversely, the information security sector surged, with stocks like Digital Certification and South Star shares hitting the daily limit, and others like Yonyou Network and Yongxin Zhicheng rising over 8% [2][4] - The assisted reproduction sector was active, with stocks like Anke Bio, Gongtong Pharmaceutical, and Lideman reaching the daily limit of 20% [11][12] Information Security Sector - The information security concept saw a strong rally, with stocks such as *ST Guandao and Digital Certification hitting their daily limits, and others like Yonyou Network and Deepin Technology also performing well [4][5] - The National Internet Information Office recently addressed Nvidia regarding security risks associated with its H20 computing chips, which may impact the information security landscape [6] AI Industry Chain - Stocks related to the AI industry chain, including liquid-cooled servers and CPO concepts, experienced significant gains, with Fangsheng Co. rising over 15% and Dingtong Technology increasing over 10% [8][9] - Domestic AI models, represented by the DeepSeek series, are now comparable to leading overseas models, with advancements in AI chip manufacturing by companies like Huawei and Haiguang Information [10] Assisted Reproductive Technology - The recent announcement of a child-rearing subsidy policy is expected to stimulate consumption in the mother and baby-related sectors, with a budget of approximately 90 billion yuan allocated for this initiative [13] - The policy is anticipated to boost birth rates and positively impact industries such as assisted reproduction, genetic testing, and childcare services [13]

Core Insights - The top conference in the field of natural language processing, ACL, awarded the best paper to a joint work by DeepSeek and Peking University titled "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention" [4][3] - This paper introduces a significant advancement in the efficiency of large language models, achieving up to 11 times faster inference while maintaining model performance [5][34] Group 1: Technology and Innovation - The paper presents a novel approach to sparse attention, moving from theoretical reasoning to a complete training process, which is crucial for the future of large models [5][26] - The Native Sparse Attention (NSA) method mimics human reading strategies by compressing long texts, selecting relevant details, and maintaining a sliding window of recent context [26][30] - NSA is designed to be natively trainable, allowing the model to learn efficient attention distribution from the pre-training phase [32][51] Group 2: Performance Metrics - In various benchmark tests, the 27B model utilizing NSA outperformed traditional full attention models in 7 out of 9 metrics, particularly excelling in reasoning tasks [35][37] - The NSA method achieved a 100% information retrieval accuracy in long text comprehension tasks, demonstrating its effectiveness in handling extensive data [38][40] - Training speed improved significantly, with forward computation accelerated by 9 times and backward propagation by 6 times, while inference speed saw an impressive 11.6 times increase [44][45] Group 3: Market Implications - The advancements in NSA technology position DeepSeek as a potential leader in the AI application ecosystem, promising faster, more efficient, and cost-effective solutions for users [55][58] - The ability to process extensive documents and datasets without manual segmentation could revolutionize how users interact with AI, enhancing productivity and accessibility [54][59] - The competitive edge provided by NSA technology is expected to solidify DeepSeek's market position, transforming it from a price-driven player to a technology innovator [58][60]

Core Viewpoint - The performance of domestic large models represented by DeepSeek has reached a level comparable to leading overseas models, with lower costs, indicating a significant improvement in usability and application potential in various sectors such as finance and office management [1][6][7]. Industry Overview - The computer industry is experiencing a fundamental improvement, with listed companies showing better performance in Q1 2025. The computer equipment sector is outperforming software development and IT services. The industry index has outperformed the CSI 300 index, ranking 6th among 31 Shenwan primary industries [3]. - The AI chip market is projected to grow from approximately $42.2 billion in 2022 to $92 billion by 2025, with a CAGR of 27.7%. Domestic AI chip manufacturers are making significant progress due to the inability to import advanced AI chips from the U.S. [4]. Algorithm and Application - Domestic large models are continuously improving, with a focus on three application areas: finance, office, and AI agents. The open-source, low-cost, and high-performance characteristics of these models are expected to accelerate their deployment in both B-end and C-end applications [3][6]. AI Computing Power - There is a strong demand for intelligent computing power, and domestic AI chip manufacturers such as Haiguang Information, Huawei, and Cambrian are achieving breakthroughs. The domestic AI chip industry is making significant strides towards self-sufficiency [4][6]. Intelligent Driving - The intelligent driving market in China, valued at trillions, is rapidly expanding, with L2-level assisted driving becoming common in mid-to-high-end models. The market is expected to see significant growth, with the total industry output value projected to increase by 7.295 billion yuan by 2025 and 25.825 billion yuan by 2030 [5][7]. Investment Recommendations - The company maintains a "stronger than the market" rating for the computer industry, highlighting investment opportunities in AIGC and intelligent driving themes. Recommended stocks include Haiguang Information, Inspur Information, and others in the AI computing power and algorithm/application sectors [2][7].

Core Insights - The continuous decrease in inference costs of large models and rapid development of domestic chips are driving the localization and deployment of inference models in data-sensitive industries such as government, finance, and healthcare [1] - The "XunTest" tool developed by China Mobile Qilu Innovation Institute addresses the challenge of efficiently and accurately testing model inference performance, which is crucial for selecting high-performance and low-cost chips [1][3] Group 1 - The "XunTest" tool innovatively constructs a "configuration equals testing" model and integrates powerful automatic data parsing capabilities, significantly improving chip selection efficiency [1][3] - The application of the "XunTest" tool has led to a remarkable transformation in efficiency, reducing the average manual monitoring time for a single test from 8 hours to 0.5 hours and decreasing data organization time by 70%, resulting in an overall chip selection efficiency improvement of 3 times [1][3] Group 2 - The core competitiveness of "XunTest" lies in two major technological highlights: intelligent automatic testing based on vLLM and automatic summarization and visualization of testing data [3] - The tool supports both local and remote testing modes, allowing flexible adaptation to different chip deployment needs, and enables engineers to initiate full-process tasks with a single configuration, overcoming the inefficiencies of traditional manual testing [3] - A standardized data storage mechanism is employed to automatically calculate and generate key performance indicators, ensuring comparability of test results across different chip platforms and heterogeneous environments [3]

Core Viewpoint - The article highlights the emergence of a new wave of entrepreneurship in the AI and robotics sectors, primarily driven by the younger generations, particularly those born in the 1990s and 2000s, who are leveraging technological advancements to create innovative solutions [1][2]. Group 1: Characteristics of the New Entrepreneurship Wave - The current entrepreneurial trend is characterized by a strong focus on AI applications and robotics, with the younger generations becoming the main driving force [2][6]. - Young entrepreneurs are adapting quickly to market changes, shifting their project directions based on real-time feedback and market demands, as seen in the case of Wang Yi who pivoted from a social app to a cash-generating AI tool [9][10]. - The average age of key teams in AI startups is notably low, with many members being under 30, indicating a generational shift in the industry [1][2]. Group 2: Global Perspective and Practicality - The new generation of entrepreneurs possesses a global outlook, engaging with international projects and communities, which enhances their innovation capabilities [12][13]. - There is a pragmatic approach to entrepreneurship, with a focus on generating cash flow quickly rather than pursuing large-scale projects that require significant upfront investment [10][11]. - The trend of smaller teams is increasing, with the proportion of teams with fewer than five members rising from 32% in 2020 to 68% in 2025, reflecting a shift towards leaner operations [11]. Group 3: Challenges and Market Dynamics - Despite the enthusiasm in the AI sector, there are significant challenges, including limited entrepreneurial directions and a competitive landscape dominated by major tech companies [15][16]. - The funding environment for AI startups is tightening, with a 16% year-over-year decline in financing events in 2024, indicating a cautious approach from investors [16]. - The rapid pace of technological advancement poses a risk for startups, as they must continuously innovate to stay relevant, with many products lacking significant technological barriers [17].

Core Viewpoint - The article highlights the emergence of a new generation of entrepreneurs, primarily from the post-90s and post-00s, who are leading the AI and robotics sectors, driven by rapid technological advancements and a favorable entrepreneurial environment [1][2][5]. Group 1: Entrepreneurial Trends - The average age of key teams in AI startups is notably young, with many members in their 20s, indicating a shift towards younger leadership in the tech industry [1][2]. - The AI boom, particularly following the launch of ChatGPT in 2022, has catalyzed a surge in entrepreneurial activity, with a focus on AI applications and robotics [1][2][5]. - Young entrepreneurs are adapting quickly to market demands, shifting their project focuses based on investor interests and market trends, as seen in the case of Wang Yi [6][7]. Group 2: Characteristics of Young Entrepreneurs - Young entrepreneurs possess a strong technical intuition and adaptability, having grown up with technology, which gives them an edge over older generations [5][6]. - Data shows that by 2025, 67% of high-potential companies in the AI sector will have founders born in the 90s, with a 120% increase in entrepreneurs from the 00s [5]. - The entrepreneurial approach of younger generations is characterized by a "light asset + agile development" model, aligning well with the rapid iteration of AI technologies [5][8]. Group 3: Global Perspective and Market Adaptation - The new generation of entrepreneurs is more globally minded, leveraging international trends and technologies, which contrasts with the more localized focus of previous generations [8][9]. - They are increasingly aware of the need for cash flow and practical business models, often pivoting their strategies to ensure sustainability and profitability [6][7][8]. Group 4: Challenges and Market Dynamics - Despite the enthusiasm, the AI sector faces significant challenges, including a competitive landscape dominated by major players, making it difficult for startups to find viable C-end opportunities [10][11]. - Funding for AI projects has seen a decline, with a 16% drop in financing events in 2024, attributed to many projects failing to demonstrate commercial viability [11][12]. - The rapid pace of technological advancement poses a risk for startups, as they must continuously innovate to stay relevant, with many products lacking significant technological barriers [10][12].