协创数据在算力方面有哪些具体表现？ 在算力方面，协创数据表现出色。今年春节期间国产 AI 大模型井喷式发展，尤 其是在推理侧实现弯道超车，与海外大厂竞争。例如字节跳动 C 站引发现象级 传播，智谱和 Minimax 分别发布 GLM 5 和 M2.5，而阿里千问 3.5 Plus 性能 媲美 Gemini 3 Pro，总参数 3,970 亿但激活仅 170 亿，以小胜大部署显存占 较前代降低 60%，推理效率提升且 API 价格降至每百万 token 0.8 元，仅为谷 歌 1/18。这些发展推动云涨价逻辑，即 AI 推理所需 token 需求年化数十倍增 长，引发产业链全面通胀，从存储到 CPU、算力租赁再到云环环相扣。 近期 H100 等老卡价格不降反升，高算以及 B 卡价格也持续上涨，这主要由于折旧 期残值重估被全世界认可。此外，国产大模型迭代导致需求暴增，例如 GLM 5 上架仅 4 天就因服务排队、响应延迟及卡顿现象而多次扩容，但依然供不应求。 这种情况下，拥有稳定、高稳定性算力供给的厂商将站在 AI 推理时代战略制高 点。 协创数据 20260223 摘要 时租赁价格显著上涨，部分平台 A100 ...

Summary of CloudWalk Technology Conference Call Company Overview - **Company**: CloudWalk Technology - **Industry**: AI and Intelligent Solutions Core Points and Arguments - **Strategic Transformation**: CloudWalk is transitioning from a tool-based company to a comprehensive solution provider, focusing on AI infrastructure and AI agents to drive customer growth [2][3] - **Dual-Driven Strategy**: The company implements a "dual-driven" strategy combining AI infrastructure and AI agents, alongside a "dual-line" development approach targeting serious industries and emerging markets [2][3] - **Technological Advancements**: Emphasis on multi-modal technology, particularly in visual models, to meet industry demands for real-time data and real-world applications [2][3] - **Intelligent Platforms**: Development of specialized platforms like Aging and Coding, structured into three layers: technology platform, business product platform, and growth/commercialization platform [2][6] - **Revenue Model Shift**: Transitioning from project-based to customized services with long-term operational contracts, enhancing revenue diversification and profitability [2][9] - **Cash Flow Stability**: Focus on reducing TOP and TOG business scales and large-scale integration projects to ensure stable cash flow [2][10] - **Customer Retention**: Over 50% repurchase rate from single clients in banking, public security, and government sectors, with long-term service contracts enhancing customer loyalty [3][18] Future Plans - **Infrastructure Development**: Plans to enhance infrastructure with national models and training grounds, aiming for broader commercialization in verticals like finance and governance from 2026 to 2028 [4][11] - **Market Expansion**: Intent to increase international market presence through high-margin, replicable delivery models, expecting significant revenue growth starting in late 2026 [12][16] - **Focus on Software Industry**: Anticipation of a software explosion period due to clearer commercialization paths and improved technical readiness [5][16] Challenges and Responses - **To B Sector Challenges**: High investment and customization demands, along with long delivery cycles, are addressed by standardizing and reusing project components to reduce costs and improve efficiency [10][13] - **Competitive Advantages**: CloudWalk holds a leading position in visual technology and has unique strengths in data understanding and real-time processing, enhancing its competitive edge [14][15] Market Outlook - **AI Industry Potential**: The software industry is expected to see significant growth, with CloudWalk positioned to leverage its technological advancements and market strategies for future success [5][12] - **Valuation Trends**: High valuations of native AI companies in the Hong Kong market reflect strong investor interest, indicating potential for CloudWalk to achieve higher valuations as it continues to innovate [18] Conclusion - **Growth Expectations**: CloudWalk anticipates double-digit revenue growth in 2026, with a projected growth rate between 20% and 40%, aiming for profitability in the near future [15][19]

人形机器人量产拐点与产业链机遇 20260223 摘要 中国人形机器人公司如丰元动力、宇树科技、魔法原子和银河通用，分 别展示了价格亲民、软硬件集成、自主可控和商业应用等不同发展方向， 预示行业多元化趋势。 海外市场对中国人形机器人在运动控制算法、硬件爆发力和自动恢复能 力方面给予高度评价，尤其对宇树科技的技术实力表示认可，认为其已 达到较高水平。 宇树科技计划 2026 年重点拓展导览和相对固定工业场景，预计出货量 达 1 万至 2 万台，产能充足，但智能模型仍是其发展的技术瓶颈，需 3 年左右时间突破。 王星星持续强调反内卷、关注智能模型、软硬件结合，并认为人形机器 人热度远未达到大模型成熟后的潜力，公司战略从技术研发向商业布局 转变。 机器人产业链对大模型的看法趋于精准，行业等待新的模型或现有模型 完善，关注点在于如何将人形机器人推向商业化，实现更大突破。 春晚人形机器人表现超出预期，短期内将提升出货量，但长期投资仍需 关注大模型和场景应用，具备场景和智能能力的公司更具投资价值。 特斯拉计划 2026 年推出第三代人形机器人 V3 产品，有望成为爆款， 建议关注拓普、三花、新建系五洲新春、恒立液压和震裕 ...

电子行业开年观点汇报 20260223 摘要 国产算力正从硬件驱动转向模型和应用驱动，字节跳动等头部互联网公 司加大自研芯片投入，计划量产 AI 推理芯片，有望占据市场份额，信源 股份等公司受益于豆包大模型需求超预期。 海外 AI 热点集中在 OpenAI 项目，其数字员工模式深度嵌入工作流，推 动推理算力需求数量级跃升，利好 Minimax、智谱 GLM 等国产模型企 业级 API 调用量增长，对 AI 芯片及基础设施是重大利好。 各大互联网公司、服务器厂商及 GPU 公司发力研发国产超级计算机方案， Switch 芯片环节受益，建议关注盛科通信等独立第三方以太网 switch 芯片厂商，以及光模块、液冷散热、PCB 等细分赛道。 2026 年一、二季度，高密度光纤连接器和高功率连续激光芯片的新需 求将成为 A 股细分赛道价值重估的核心变量，英伟达新一代 AI 芯片的曝 光也将为 GPU 的发展指引新的增长方向。 英伟达最新旗舰产品 120.4T 交换机将推动高端 PCB、CCL 材料及上游 核心元件供应商发展，相关厂商如胜宏、生益、菲利华等有望受益， HVLP 铜箔和树脂供应商可能面临供不应求。 Q&A ...

Summary of Key Points from Conference Call Industry Overview - The conference call primarily discusses the **AI**, **CPU**, and **optical fiber communication** industries, highlighting the performance of companies like **Lumentum**, **长飞 (Changfei)**, and **亨通 (Hengtong)** during the 2026 Spring Festival period [1][2]. Core Insights and Arguments - **AI and Agent Technology**: The application of Agent technology is driving demand for personal terminal devices, with significant updates from major players like **字节跳动 (ByteDance)** and **阿里巴巴 (Alibaba)**, as well as overseas models like **GPT-3.1** and **Cloud 4.6** [2]. - **CPU Market Performance**: Companies in the CPU sector, particularly **Lumentum**, have shown strong performance despite challenges in the market, with expectations for rationality as new product launches approach [2]. - **Optical Fiber Communication Growth**: Companies like **长飞** and **亨通** are benefiting from domestic infrastructure projects and global division of labor, with high-end product exports contributing to growth potential [1][2]. - **CPO Technology Impact**: CPO technology has a limited impact on power consumption reduction (2%-4%), and the stability of AI infrastructure is deemed crucial for large capital expenditure projects [7][8]. - **NPU Adoption**: The NPU (Network Processing Unit) solution is expected to gain traction, potentially increasing shipment volumes significantly by 2027 [9]. Emerging Trends - **Cabinet Market Dynamics**: The optical communication field is seeing a shift towards low-power, high-efficiency optical interconnection solutions, with CPU and NPU becoming mainstream choices [3]. - **Silicon Photonics Development**: Companies like **新易盛 (NewEase)** are advancing in silicon photonics, with the potential for large-scale production of core components [11][12]. - **Challenges in Silicon-Based Technology**: The industry faces challenges related to light emission, heat control, and signal modulation, necessitating collaboration across the supply chain [13]. Market Opportunities - The optical communication market is viewed as a growth opportunity for both established and emerging companies, with various regions (Mainland China, Taiwan, North America) having unique advantages [5]. - A-share companies in the optical communication sector are considered to have investment value, particularly in the context of an expanding market [6]. Additional Insights - **AI Application Penetration**: Currently, only about 0.01% of the population is paying for AI services, but this is expected to grow rapidly with advancements in model capabilities [16]. - **User Behavior Changes**: There is a notable shift in user behavior from app engagement to API interactions via agents, leading to increased uncertainty in software ecosystems [17]. - **Domestic Inference Computing Limitations**: Domestic inference computing is currently insufficient to meet the demand for agent overflow, highlighting a need for improved infrastructure [19]. - **Future AI Trends**: The proliferation of personal agents is anticipated to lead to exponential growth in token consumption and hardware demand, particularly for GPUs [20][21].

格隆汇2月24日丨金现代(300830.SZ)在互动平台表示，公司的轻骑兵低代码开发平台已经应用了 DeepSeek等大模型。目前在积极关注DeepSeekV4等版本的新特性及其发布进度等。 （原标题：金现代(300830.SZ)：轻骑兵低代码开发平台已经应用了DeepSeek等大模型） ...

Core Viewpoint - The company, Rockchip (瑞芯微), is actively promoting its RK3588 and RK182X chips for personal deployment of AI Agent applications, indicating a strategic focus on enhancing user data privacy and low-latency responses [1] Group 1: Product Capabilities - The RK3588 chip is already being utilized in AI Agent application solutions that leverage cloud-based large models [1] - The RK182X co-processor offers high computing power and bandwidth, enabling various smart terminal devices to deploy large models locally [1] Group 2: Market Demand - The solutions provided by the RK182X are designed to meet user demands for data privacy, security, and efficient multi-tasking capabilities [1]

2026 年 2 月，国际市场调研机构沙利文（Frost & Sullivan）发布《中国 GenAI 市场洞察：企业级大模型调用全景研究，2025 H2》报告显示：2025 年下半 年，中国企业级大模型日均调用量飙升至 37.0 万亿 tokens，较上半年的 10.2 万亿增长 263%。 头部大模型占比均有提升，其中，阿里云千问（Qwen）增幅最多，占比跃升至 32.1%，相较上半年的 17.7% 几乎翻倍，领先优势扩大，成为最受中国企 业青睐的大模型。 2025 年中国企业级大模型日均调用量： 2025 年是大模型从实验室走向企业生产的关键一年。 基于对 870 份来自互联网、金融、消费电子及汽车等重点行业有效问卷的分析，此次研究发现，企业应用大模型的动因发生了本质变化。 2025 年上半年，企业使用大模型的主要驱动力是提升产品性能与客户体验；到下半年，提升企业运营效率和研发效能跃升为首要动因，大模型也成为企 业应对市场竞争压力的关键工具。 在此背景下，企业级大模型市场正加速向头部厂商集中。可以 2025 年下半年，大模型日均 tokens 消耗 TOP 3为：千问（32.1%）、豆包（21.3%）和 ...

距团队正式组建扩张仅半年，无界动力已在技术研发和商业化等方面取得了多项关键进展。在算法模型 层面，无界动力坚持投入具身智能关键技术创新，系统性构建长期技术壁垒。为持续突破操作智能在动 态物理世界完成高复杂任务的性能上限和泛化性的提升，无界动力构建了内嵌隐式世界模型的端到端多 模态大模型，通过数据驱动的高维表征内化环境动态与物理规律，使机器人不仅具备持续进化的3D感 知理解、历史记忆建模、未来环境变化预测和细腻动作生成的能力，也同时构建和提升反思、纠错等高 阶能力，实现系统层级性能和鲁棒性的显著提升。目前，该模型已在工业装配、柔性物体操作、清洁整 理等多项任务上实现泛化性与成功率的同步突破，获多家客户认可。 2026年是"十五五规划"开局之年，是"人工智能+"为代表的新质生产力全面发力的一年。全球正处于百 年未有之大变局，以AI为代表的新一轮科技革命正在重塑全球产业格局，具身智能、大模型、智能网 联等前沿技术加速突破，为全球经济增长注入新动能，为企业发展和财富增长提供新机遇。 近日，通用具身智能机器人公司无界动力宣布完成超2亿元天使+轮融资，老股东红杉中国、线性资 本、高瓴创投、华业天成、中金资本旗下基金、BV ...

Core Viewpoint - The A-share market experienced a strong opening on the first trading day after the Lunar New Year, with major indices showing positive performance, particularly in resource and hard technology sectors [1][2][4]. Market Performance - On February 24, the Shanghai Composite Index rose by 0.87% to 4117.41 points, while the Shenzhen Component increased by over 1% and the ChiNext Index gained 0.99%. The total market turnover exceeded 2 trillion yuan, with over 4000 stocks rising and 111 stocks hitting the daily limit [1][2][3]. - The market saw a strong performance in defensive sectors, with resource stocks experiencing a rebound. Key sectors such as energy equipment, oil and gas, and precious metals led the gains [1][3][4]. Sector Analysis - The hard technology sector also showed significant strength, particularly in optical modules (CPO) and optical communication concepts, which saw notable increases. In contrast, several large model concepts, including AI applications and the film and media sector, experienced declines [1][4]. - The human-robot concept stocks opened high, reflecting investor interest in this emerging technology [2][3]. Analyst Insights - Analysts suggest that the market's overall performance is promising, with resource stocks providing substantial support to the indices. The Shenzhen Component rose by 1.36%, indicating active growth in certain sectors [4]. - Geopolitical risks and uncertainties in global trade policies are seen as short-term disturbances, but they do not alter the mid-term trends. The expectation of a stronger renminbi and upcoming policy influences from the "Two Sessions" are expected to support the market [6][7]. Investment Strategies - Analysts recommend focusing on structural opportunities rather than broad index movements. Key investment themes include sectors benefiting from improved supply-demand dynamics and industry profit recovery, such as non-ferrous metals, basic chemicals, and construction materials [7][8]. - The robot sector is highlighted as having dual support from policy expectations and capital inflows, with a strong performance anticipated due to ongoing demand for new productive forces and smart manufacturing [8].