金融界消息 截至2025年7月29日收盘，云天励飞U（688343）报收于54.97元，下跌0.11%，换手率 3.56%，成交量9.38万手，成交金额5.14亿元。 资金流向方面，今日主力资金净流出4399.19万元，占比成交额8.56%。其中，超大单净流出3628.31万 元、占成交额7.06%，大单净流出770.89万元、占成交额1.5%，中单净流出流入1612.63万元、占成交额 3.14%，小单净流入2786.57万元、占成交额5.42%。 云天励飞最新一期业绩显示，截至2025一季报，公司营业总收入2.64亿元、同比增长168.23%，归属净 利润8564.20万元，同比增长37.86%，扣非净利润11095.86万元，同比增长29.21%，流动比率5.468、速 动比率5.060、资产负债率32.59%。 天眼查商业履历信息显示，深圳云天励飞技术股份有限公司，成立于2014年，位于深圳市，是一家以从 事软件和信息技术服务业为主的企业。企业注册资本35513.372万人民币，实缴资本8367.208万人民币。 公司法定代表人为陈宁。 通过天眼查大数据分析，深圳云天励飞技术股份有限公司共对外投资了3 ...

Core Viewpoint - The company Yuntian Lifei announced its commitment to AI chips during the 2025 WAIC, marking a significant push for domestic computing power in the AI era [2] Group 1: AI Chip Strategy - Yuntian Lifei will focus on three core areas: edge computing, cloud large model inference, and embodied intelligence, aiming to create a cost-effective "computing power accelerator" for the AI age [2] - The CEO, Chen Ning, emphasized that 2025 will be a pivotal year for AI, as large model technology matures and costs decrease, transforming AI from an "expert tool" to a "universal infrastructure" [2][3] - The market demand for inference computing power is expected to experience explosive growth as AI transitions from training to inference [2] Group 2: Market Potential and Growth - Although the current market for AI inference chips is relatively small compared to the multi-hundred billion dollar training chip market, it is projected to grow significantly faster in the next 3-5 years [2][3] - Chen Ning compared the relationship between training and inference to electricity generation and consumption, suggesting that inference chips will become ubiquitous across various devices [3] Group 3: Financial Performance and Future Plans - Yuntian Lifei plans to pursue a secondary listing in Hong Kong to support its AI strategy, enhance financing channels, and facilitate international market expansion [3] - The company anticipates revenue exceeding 900 million yuan in 2024, representing an 81.3% year-on-year increase, with Q1 2025 revenue expected to grow over 160% year-on-year [3] Group 4: Technological Development - Since its establishment in 2014, Yuntian Lifei has focused on the practical application of AI technology, particularly in reducing AI computing costs through chip development [3][4] - The company has launched five generations of NPU chips, establishing a product matrix that covers edge inference, large model inference, and embodied intelligence [4]

Core Insights - The rapid development of AI Agents is increasing the importance of inference chips, with major computing manufacturers showcasing new products at the WAIC 2025 [1] - Cloud Tianli Fei announced a focus on AI chips, aiming to build a domestic computing "accelerator" around three core areas: edge computing, cloud large model inference, and embodied intelligence [1] - The chairman and CEO of Cloud Tianli Fei, Chen Ning, believes that AI technology centered on large models and inference chips will redefine all electronic products in the next five years [1][2] Inference Chip Market Dynamics - Chen Ning compares the evolution of AI to a student graduating from university, indicating that the current phase is transitioning from an AI training era to an inference era [2] - The inference era will see AI empowering all electronic products, necessitating various types of inference chips from terminals to edge and cloud computing [2] - Chen Ning emphasizes the need for a collaborative approach between edge, cloud, and terminal computing to optimize cost-effectiveness in large-scale deployments [2][3] Cost Considerations in Inference Chips - In the transition to the inference era, the cost of inference will become increasingly important as AI becomes integrated into daily life [3] - The core concept of PPA (Performance, Power consumption, Area) is crucial in chip design, influencing the value and cost-effectiveness of chips from the user's perspective [3][4] - For edge computing, the balance between computational power, memory, and customized services is essential, with effective power and hardware cost being key factors [3][4] Performance Metrics for Cloud and Edge Inference - Cloud-side inference focuses on the hardware procurement costs and operational expenses of running inference chip clusters, while edge computing is more sensitive to the hardware costs and effective computational power in specific scenarios [4] - The effective computational power and hardware cost are critical in assessing the cost-performance ratio of hardware devices [4] - Cloud Tianli Fei is concentrating on building a high-cost performance inference chip technology and product system to drive the large-scale deployment of AI applications [4]

Core Insights - AI inference chips are emerging as a new focus in the artificial intelligence industry, with Shenzhen Yuntian Lifeng (688343.SH) announcing a comprehensive focus on this area during the World Artificial Intelligence Conference in 2025 [1][2] - The CEO of Yuntian Lifeng, Chen Ning, highlighted that 2025 will be a pivotal year for AI development, with significant reductions in model invocation costs and a shift from AI as an "expert tool" to a "universal infrastructure" [1][2] - The demand for inference computing power is expected to experience explosive growth as AI transitions from training to inference [1][3] Industry Trends - The report from CITIC Securities indicates that three main factors are accelerating the demand for inference computing power: the integration of AI with existing internet businesses, the combination of agents and deep reasoning, and the penetration of multimodal capabilities [2] - AI is anticipated to redefine various electronic products, including wearable devices and household appliances, enabling them to interact more naturally and respond to complex commands [2] Company Developments - Yuntian Lifeng is focusing on AI inference chips, which are categorized into training chips and inference chips, with the latter being crucial for utilizing neural network models for predictions [3] - The company has developed four models of chips: DeepEdge10C, DeepEdge10 Standard, DeepEdge10Max, and DeepEdge200, with the DeepEdge10 series specifically designed for edge AI applications [3][4] - The DeepEdge10 series employs a "computing power building block" architecture, allowing for scalable integration of computing units to meet varying power requirements [4][5] Financial Performance - Yuntian Lifeng reported an 81% revenue growth in 2024, with a further increase to 160% in the first quarter of this year [5] - The management expressed confidence in maintaining high growth rates in the second half of the year, driven by advancements in AI inference algorithms and increasing demand for computing power [5]

北京市金杜（深圳）律师事务所 关于深圳云天励飞技术股份有限公司 之法律意见书 致：深圳云天励飞技术股份有限公司 北京市金杜（深圳）律师事务所（以下简称本所）接受深圳云天励飞技术股 份有限公司（以下简称公司）委托，根据《中华人民共和国证券法》（以下简称《证 券法》）、《中华人民共和国公司法》（以下简称《公司法》）、中国证券监督管理委 员会（以下简称中国证监会）《上市公司股东会规则》（以下简称《股东会规则》） 等中华人民共和国境内（以下简称中国境内，为本法律意见书之目的，不包括中 国香港特别行政区、中国澳门特别行政区和中国台湾省）现行有效的法律、行政 法规、规章、规范性文件和现行有效的公司章程有关规定，指派律师出席了公司 于 2025 年 7 月 28 日召开的 2025 年第二次临时股东大会（以下简称本次股东大 会），并就本次股东大会相关事项出具本法律意见书。 为出具本法律意见书，本所律师审查了公司提供的以下文件，包括但不限于： 份有限公司章程》（以下简称《公司章程》）； 圳云天励飞技术股份有限公司第二届董事会第十六次会议决议公告》 《深圳云天励 飞技术股份有限公司第二届监事会第十四次会议决议公告》； 圳云天励 ...

案》 审议结果：通过 | 证券代码：688343 公告编号：2025-045 证券简称：云天励飞 | | | | | | | --- | --- | --- | --- | --- | --- | | 深圳云天励飞技术股份有限公司 | | | | | | | 本公司董事会及全体董事保证公告内容不存在任何虚假记载、误导性陈述或 | | | | | | | 者重大遗漏，并对其内容的真实性、准确性和完整性依法承担法律责任。 | | | | | | | 重要内容提示： | | | | | | | ? 本次会议是否有被否决议案：无 | | | | | | | 一、 会议召开和出席情况 | | | | | | | （一） 股东大会召开的时间：2025 年 7 | 月 28 日 | | | | | | （二） 股东大会召开的地点：广东省深圳市南山区粤海街道深圳湾科技生态园 | | | | | | | （三） 出席会议的普通股股东、特别表决权股东、恢复表决权的优先股股东及 | | | | | | | 其持有表决权数量的情况： | | | | | | | 普通股股东人数 238 | | | | | | | 普通股股东所持有表决 ...

Group 1 - The year 2023 is marked as the year of AI technology application, with the World Artificial Intelligence Conference (WAIC) showcasing groundbreaking innovations, particularly in chip and computing power development [2] - Guangdong-based company Yuntian Lifeng presented its self-developed DeepEdge10 series chips, featuring a "computing power building block" architecture that allows flexible assembly and expansion of computing power [2] - Yuntian Lifeng is strategically focusing on AI inference chips this year, planning to build a domestic computing power "accelerator" around three core areas: edge computing, cloud-based large model inference, and embodied intelligence [2] Group 2 - Dr. Chen Ning, Chairman and CEO of Yuntian Lifeng, stated that 2025 will be a pivotal year for AI development, with large model technology reaching new maturity and significantly reduced model invocation costs, transitioning AI from an "expert tool" to a "universal infrastructure" [2] - The demand for inference computing power is expected to experience explosive growth as AI transitions from a training era to an inference era [2] - AI is anticipated to reshape various electronic products, including wearable devices, household appliances, and electric vehicles, redefining their forms and functions [3] Group 3 - The underlying support for these transformations relies on AI inference chips, which will create a ubiquitous computing power network across endpoints, edges, and clouds [3] - This "full coverage" computing foundation enables conversational AI to operate efficiently across various devices, facilitating the transition of electronic products from "tools" to "intelligent partners" [3]

深圳云天励飞技术股份有限公司 关于选举第二届董事会职工代表董事的公告 证券代码：688343 证券简称：云天励飞 公告编号：2025-046 特此公告。 深圳云天励飞技术股份有限公司董事会 2025 年 7 月 29 日 附件： 李建文先生简历 李建文，1984 年出生，中国国籍，无境外永久居留权，2009 年毕业于西北 工业大学，获得硕士学位；2009 年 5 月至 2011 年 9 月，在中兴通讯担任芯片开 发经理；2011 年 9 月至 2013 年 10 月，自主创业；2013 年 11 月至 2025 年 4 月， 在深圳图灵微担任副总经理；2014 年 8 月加入公司，现任公司董事、运营体系 总经理。 本公司董事会及全体董事保证本公告内容不存在任何虚假记载、误导性陈 述或者重大遗漏，并对其内容的真实性、准确性和完整性依法承担法律责任。 深圳云天励飞技术股份有限公司（以下简称"公司"）分别于 2025 年 7 月 11 日、2025 年 7 月 28 日召开第二届董事会第十六次会议和 2025 年第二次临时 股东大会，审议通过了《关于取消监事会并修订<公司章程>及相关议事规则的议 案》，修订后的《深 ...

北京市金杜（深圳）律师事务所 关于深圳云天励飞技术股份有限公司 2025 年第二次临时股东大会 之法律意见书 致：深圳云天励飞技术股份有限公司 北京市金杜（深圳）律师事务所（以下简称本所）接受深圳云天励飞技术股 份有限公司（以下简称公司）委托，根据《中华人民共和国证券法》（以下简称《证 券法》）、《中华人民共和国公司法》（以下简称《公司法》）、中国证券监督管理委 员会（以下简称中国证监会）《上市公司股东会规则》（以下简称《股东会规则》） 等中华人民共和国境内（以下简称中国境内，为本法律意见书之目的，不包括中 国香港特别行政区、中国澳门特别行政区和中国台湾省）现行有效的法律、行政 法规、规章、规范性文件和现行有效的公司章程有关规定，指派律师出席了公司 于 2025 年 7 月 28 日召开的 2025 年第二次临时股东大会（以下简称本次股东大 会），并就本次股东大会相关事项出具本法律意见书。 为出具本法律意见书，本所律师审查了公司提供的以下文件，包括但不限于： 1. 经公司 2023 年第三次临时股东大会审议通过的《深圳云天励飞技术股 份有限公司章程》（以下简称《公司章程》）； 2. 公司 2025 年 7 月 ...

证券代码：688343 证券简称：云天励飞 公告编号：2025-045 者重大遗漏，并对其内容的真实性、准确性和完整性依法承担法律责任。 （三） 出席会议的普通股股东、特别表决权股东、恢复表决权的优先股股东及 其持有表决权数量的情况： 重要内容提示： 本次会议是否有被否决议案：无 一、 会议召开和出席情况 （一） 股东大会召开的时间：2025 年 7 月 28 日 （二） 股东大会召开的地点：广东省深圳市南山区粤海街道深圳湾科技生态园 10 栋 A 座 3 楼会议室 深圳云天励飞技术股份有限公司 2025年第二次临时股东大会决议公告 本公司董事会及全体董事保证公告内容不存在任何虚假记载、误导性陈述或 | 1、出席会议的股东和代理人人数 | 238 | | --- | --- | | 普通股股东人数 | 238 | | 2、出席会议的股东所持有的表决权数量 | 115,500,554 | | 普通股股东所持有表决权数量 | 115,500,554 | | 3、出席会议的股东所持有表决权数量占公司表决权数量的比例 | 32.3174 | | （%） | | | 普通股股东所持有表决权数量占公司表决权数量的比例（% ...