通信行业周报（2 月 23 日- 2 月 28 日） 通信行业市场回顾 市场研究部 2026 年 3 月 2 日 2026 年 2 月 23 日至 2 月 28 日，通信（中信）板块上涨了 4.5%，而 沪深 300 指数上涨了 1.08%，通信板块跑赢沪深 300 指数 3.43 个百 分点。期间通信业区间涨幅位列中信一级行业第 9 位，在 30 个中信一 级行业中表现中上。2026 年全年，通信行业累计上涨 5.08%，在中信 一级行业中排第 17 位，表现居中。截至 2 月 27 日，中信通信行业 PE TTM 为 27.58 倍，处于 38.63%的分位数。 周度关注：算力互联互通节点建设 2026 年 2 月 6 日，工业和信息化部办公厅正式发布《关于组织开展国 家算力互联互通节点建设工作的通知》，该通知是落实《算力互联互通 行动计划》（2025 年 5 月）的具体举措，标志着我国算力基础设施建 设从"单点突破"迈向"全网协同"的关键阶段。 过去几年，我国在"东数西算"工程推动下，算力规模迅速扩张。通过 构建"1+M+N"的节点体系，打破算力孤岛，实现全国算力资源的标 准化互联与高效调度。其中"1" ...

相关产品：人工智能AIETF（515070）、华夏中证人工智能主题ETF联接A（008585）、华夏中证 人工智能主题ETF联接C（008586） 每日经济新闻 今日A股开盘后人工智能AIETF（515070）震荡为主。截至10:25，人工智能AIETF（515070）跌 0.05%，持仓股星宸科技涨超13%，润泽科技涨超10%，光迅科技、光环新网等个股上涨明显。人工智 能AIETF（515070）成交额突破9000万元。 消息方面，2月25日，Kimi母公司北京月之暗面科技有限公司完成总额超7亿美元的新一轮融资， 由阿里、腾讯等互联网巨头领投，融资后公司估值升至100亿至120亿美元，刷新中国大模型独角兽估值 纪录。Kimi K2.5模型发布不到20天，收入已超过公司2025年全年总和，凭借长文本处理优势与高用户 黏性，月之暗面创下国内AI独角兽最快成长速度。 中邮证券分析指出，中国AI大模型进入产品成熟期，低成本高性能技术为市场增长提供支撑，多 模态模型拓展应用场景，厂商完善产品矩阵降低接入成本。企业AI需求从概念验证转向规模化落地， 带动MaaS及大模型解决方案进入结构性放量周期。 人工智能AIETF（ ...

据IPO早知道消息，在完成上一轮5亿美金融资仅一个多月后，Kimi母公司月之暗面（Moonshot AI）新一轮超7亿美元融资即将完成全部交割并实现超 募。本轮融资由阿里、腾讯、五源、九安等联合领投，高榕创投等老股东继续加码，投后估值超100亿美元。 值得一提的是，Kimi连续两轮超12亿美金的融资，创下了近一年来大模型行业的最高融资纪录。事实上，在这一轮融资即将完成之际，Kimi已经以100 亿-120亿美元的估值开启了新一轮融资，并已获得多家机构的意向，不乏早前鲜少布局中国科技企业的欧洲背景机构。 更值得关注的是，K2.5大模型发布不到一个月后，Kimi近20天的累计收入已超过2025年全年总收入，增长主要受全球付费用户及API调用量大涨共同驱 动；且Kimi的收入在K2.5发布之后出现结构性反转，海外收入已超过国内收入。 短期不着急上市、将主动权掌握自己手中，更希望在一级市探索技术与资本新型关系 不可否认的是，当下Kimi在一级市场的热度之高在一定程度上的确得益于智谱和MiniMax在二级市场屡创新高，曾一度双双市值突破3000亿港元，投资人 的FOMO（Fear of Missing Out，错失恐惧 ...

2月17日，南都N视频记者了解获悉，在完成上一轮5亿美元融资仅一个多月后，大模型独角兽月之暗面 Kimi新一轮超7亿美元（约为人民币48.36亿）的融资即将完成交割，本轮由阿里、五源、九安等老股东 联合领投，腾讯也参与投资。此外，其新一轮100亿-120亿美元估值的融资亦已经开启。 去年12月31日，月之暗面创始人、CEO杨植麟发布内部信，称完成近期的5亿美元（约为人民币34.54 亿）C轮融资后，公司账上还有超过100亿元的人民币现金储备。 杨植麟还强调，"相比二级市场，我们判断还可以从一级市场募集更大量资金，事实上，我们B/C轮融 资金额就超过绝大部分IPO募资及上市公司的定向增发。所以我们短期不着急上市。当然未来我们计划 将上市作为手段来加速AGI，择时而动，主动权掌握在我们手中。" 1月27日，月之暗面Kimi发布并开源其Kimi K2.5模型，并称它是Kimi迄今最智能的模型，该模型除了在 长文本上展示了一定的优势以外，也在Agent、代码、图像、视频及一系列通用智能任务上取得一定的 表现。自此，月之暗面kimi正式合上多模态模型的"拼图"。 本文自南都N视频 采写 | 南都N视频记者 林文琪 当智 ...

【导读】Kimi估值提升至100亿美元 中国基金报记者 泰勒 大家好，关注一则中国AI的利好消息。 审核：木鱼 2月17日，国内AI初创公司月之暗面（Moonshot AI） 正在推进一轮由阿里巴巴和腾讯支持的增额融 资，目标估值达100亿美元。 据报道，这家Kimi聊天机器人背后的公司在1月下旬开启了额外融资讨论，以满足投资者的需求。而就 在一个多月前，该公司刚以43亿美元的估值获得了5亿美元融资。照此计算，月之暗面的估值已经暴涨 超100%。知情人士表示，包括阿里巴巴、腾讯和五源资本在内的现有股东已为本轮融资的首批资金注 资超过7亿美元。 月之暗面融资速度之快，反映了投资者正急于押注一批试图在世界级AI服务领域挑战OpenAI和 Anthropic的中国初创公司。上个月，月之暗面发布了Kimi K2.5模型，在关键的春节长假前领跑了中国 主流AI模型的升级浪潮。 在分发平台OpenRouter上，K2.5是使用率最高的大语言模型之一，大幅领先于DeepSeek和谷歌的 Gemini。 在基准测试网站Artificial Analysis的开源模型排名中，K2.5目前位列第二，仅次于智谱AI最新发布的 GL ...

（来源：千龙网） 马年春节前夕，国产大模型集体"炸场"。 继豆包Seedance2.0刷屏后，2月11日，位于北京海淀的大模型企业智谱，发布了新一代旗舰模型GLM-5。同时证实，此前在海外引爆技术圈的匿名模 型"Pony Alpha"，正是GLM-5的内测版。 DeepSeek-OCR 2模型、Kimi K2.5模型、MiniMax Music2.5音乐生成模型……半个月，近十款模型落地，从通用对话到编程、音乐生成、经营决策，各有 各的"杀手锏"。这些亮眼成绩，也让北京的AI创新实力得到国际市场与科研领域的双重认可。 两年前，我们还在为文心一言画出一只像样的猫而惊叹。而今天，AI大模型已经在加速迈过从"秀肌肉"到"真干活"的临界点，突破代码和工程之间的鸿 沟。 比如，在智谱AI编程平台Z Code上，用户只需把要求说清楚，模型就能自动拆解任务，做出复杂应用；有人用手机，就可远程指挥电脑桌面的智能体， 完成了过去需要几天耗时的工程任务；在Vending Bench 2测试里，GLM-5独自经营的模拟自动售货机，一年后账户余额4432美元，这个表现已经接近了国 际顶尖模型。 必须承认，美国在算力上的优势仍是客观现 ...

Core Insights - The Chinese AI industry is experiencing an unprecedented wave of flagship model releases, with companies racing to convert technological capabilities into scalable consumer products [1][4] Group 1: Recent Developments in AI Models - DeepSeek has emerged as a strong competitor, surpassing all major AI applications in active user numbers during the 2025 Spring Festival [1] - On February 11, Zhiyuan launched its new flagship model GLM-5, focusing on programming and agent capabilities, achieving state-of-the-art performance in coding tasks [2] - MiniMax introduced its flagship programming model MiniMax M2.5 on February 12, designed for agent scenarios and achieving high performance in various productivity applications [3] Group 2: Market Reactions and Financial Performance - Following the announcements of new models, Zhiyuan's stock price increased by 28.68% and MiniMax's by 14.62%, with market capitalizations reaching 179.229 billion HKD and 184.418 billion HKD respectively [3] - Morgan Stanley's recent report rated both Zhiyuan and MiniMax as "overweight," highlighting their potential to capture the next wave of global AI value creation [4] Group 3: Competitive Landscape and Technological Advancements - The focus of AI model development has shifted from parameter scale to practical capability upgrades, with several companies releasing or upgrading models since late January [5][6] - Alibaba's Qwen3-Max-Thinking model, launched on January 26, boasts over one trillion parameters and significant improvements in various dimensions [5] - The introduction of Kimi K2.5 by Moonlight and the updates from DeepSeek and ByteDance indicate a trend towards enhancing multi-modal capabilities and increasing model capacity [6] Group 4: Future Trends and Industry Outlook - The arrival of the agent era is reshaping the internet value chain, moving from traditional traffic scale to a focus on execution capabilities and result-based business models [7] - The new value system emphasizes cloud platforms, computational resources, governance tools, and content monetization as primary profit drivers [7]

Core Viewpoint - The article discusses the rapid advancements in artificial intelligence (AI) and the intense competition between domestic and international companies, highlighting the differences in focus between the two markets [3][4]. Group 1: Recent Developments in AI - Meta has released a new model codenamed "Avocado," improving computational efficiency by 10 to 100 times [3]. - Google has opened its world model Genie 3, allowing users to create and explore 3D virtual worlds through natural language descriptions [3]. - Domestic companies like Alibaba and ByteDance are also making significant strides, with Alibaba launching a model with over one trillion parameters [3]. Group 2: Competition Dynamics - The competition in AI has reached a fever pitch, with major tech companies investing heavily, including a reported total capital expenditure of approximately $650 billion from the four major U.S. tech giants for 2026 [3]. - The "AI red envelope war" in China has seen tech giants invest over 4 billion yuan to capture the "AI super entrance" [4]. Group 3: Differences in Focus - International companies prioritize breakthroughs in underlying technology and model capabilities, while domestic firms are more focused on application and commercial realization [4][5]. - Domestic companies excel at transforming new technologies into commercial successes but often lag in original innovation [5][6]. Group 4: Challenges and Opportunities - The disparity in innovation focus is linked to differences in market foundations, technological reserves, and business environments between China and the U.S. [6]. - Despite significant investments in foundational technologies, domestic firms still face challenges in original innovation and product iteration speed compared to their international counterparts [6]. Group 5: Vision for the Future - The article emphasizes the need for Chinese tech companies to shift their focus towards original innovation to avoid being outpaced by international competitors [7]. - It calls for a cultural shift towards innovation and greater support for foundational research from leading tech firms to achieve the national goal of technological self-reliance [7].

Core Insights - The article emphasizes the importance of original innovation alongside commercial success in the tech industry, particularly in the context of rapid advancements in artificial intelligence (AI) [1][5]. Group 1: AI Developments - Major tech companies are making significant strides in AI, with Meta's new model improving computational efficiency by 10 to 100 times, and Google's Genie 3 enabling the creation of 3D virtual worlds through natural language [1]. - Domestic companies like Alibaba and Tencent are also heavily investing in AI, with Alibaba releasing a model with over one trillion parameters and Tencent planning substantial AI investments [1][2]. Group 2: Competitive Landscape - The competition in AI has intensified, with a notable "AI red envelope war" in China, where tech giants are investing over 4 billion yuan to capture the "AI super entrance" [2]. - The article contrasts the focus of U.S. companies on foundational technology breakthroughs with Chinese firms' emphasis on application and commercialization [2][3]. Group 3: Innovation and Market Dynamics - There is a notable difference in innovation strategies, with Chinese companies excelling in commercial success but lagging in original innovation compared to their foreign counterparts [3][4]. - The article suggests that the current focus on immediate commercial success may hinder long-term innovation capabilities, warning that neglecting original innovation could lead to vulnerabilities in the future [4][5]. Group 4: Call for Change - The article calls for Chinese tech companies to prioritize foundational research and innovation, highlighting that the lack of ambition and mission-driven focus may be a significant barrier to achieving technological independence [5]. - It stresses the need for a cultural shift towards innovation and the role of government in guiding and supporting these efforts to realize the dream of technological self-reliance [5].

Core Insights - The article highlights significant developments in the AI industry, including mergers, funding rounds, and technological advancements, reflecting the rapid growth and competitive landscape of AI companies globally. Group 1: Mergers and Acquisitions - SpaceX has acquired xAI, marking a strategic move by Elon Musk to integrate AI into its aerospace operations [7] - OpenAI is prioritizing the commercialization of ChatGPT, leading to the departure of several executives amid a shift in focus from long-term research to immediate product development [9] Group 2: Funding and Investments - A total of 34 AI financing events were disclosed globally, with a total funding amount of 169.13 billion RMB, averaging 8.05 billion RMB per event [42] - The highest funding in the domestic AI sector was achieved by Zhijidongli, which completed a 14.01 billion RMB A++ round of financing [49] Group 3: Technological Advancements - The first AI percutaneous navigation robot has been approved for market release in China, enhancing precision in medical procedures [9] - Tencent's HPC-Ops has improved inference throughput by 30%, showcasing advancements in AI model performance [23] Group 4: Market Trends and Predictions - The domestic transformer market is expected to grow over 20% year-on-year by 2025, driven by the increasing demand for AI computing power [40] - The CEO of Xiaopeng Motors predicts that AI will be a central theme in technology for the next 30 to 50 years, emphasizing its integration into the automotive industry [38] Group 5: Company Strategies and Developments - Ant Group's CEO announced an "AI Credit" incentive program to encourage innovative contributions in AI [29] - Meitu's CEO discussed the competitive landscape between general AI models and specialized applications, emphasizing the importance of niche markets [19]