消息面上，博泰车联近日宣布，与平安财险订立框架合作协议，双方将通过"AI+车联网技术+保险服 务"的深度融合创新，打破汽车产业与金融保险产业的行业边界，实现从"基础服务"向"个性化、全场景 赋能"的跨越式升级，重构智慧出行保险生态，引领智慧出行时代的保险服务变革与用户体验升级。 值得注意的是，依托其在智能座舱、大模型及具身智能等前沿领域的深厚积累，博泰车联正加速推动技 术成果向商业化场景转化。2025年10月，公司与人工智能领军企业商汤科技达成全面战略合作，聚焦 AI大模型在智能座舱中的深度应用，并同步布局人形机器人与舱驾一体等未来赛道，彰显其向高阶智 能出行生态跃迁的决心。 博泰车联 分时图 日K线 周K线 月K线 280.00 10.20 3.78% 4.37% 2.92% 1.46% 0.00% 1.46% 2.92% 4.37% 258.00 261.93 265.87 269.80 273.73 277.67 281.60 09:30 10:30 12:00/13:00 14:00 16:10 0 733 1467 2200 博泰车联(02899)午后涨超4%，高见282港元，再创上市新高。截至发稿 ...

Core Viewpoint - Botai Carlink (02899) has seen a significant increase in stock price, reaching a new high of 282 HKD, driven by a strategic partnership with Ping An Insurance to innovate insurance services through AI and connected vehicle technology [1] Group 1: Company Developments - Botai Carlink's stock rose over 4% in the afternoon session, with a trading volume of 10.42 million HKD [1] - The company announced a framework cooperation agreement with Ping An Insurance, aiming to integrate AI, connected vehicle technology, and insurance services to enhance user experience and transform the insurance ecosystem [1] Group 2: Technological Advancements - Botai Carlink is accelerating the commercialization of its technological achievements in smart cockpits, large models, and embodied intelligence [1] - A comprehensive strategic partnership was established with SenseTime in October 2025, focusing on the deep application of AI large models in smart cockpits and exploring future avenues such as humanoid robots and integrated cockpit-driving systems [1]

Investment Rating - The report maintains a "Buy" rating for several companies in the media sector, including Giant Network, Perfect World, and Kuaishou, among others [5][36]. Core Insights - The media industry showed a slight increase of 0.21% from January 26 to February 1, 2026, outperforming the Shanghai Composite Index, which decreased by 0.44% [12]. - The advertising and marketing sub-sector led the gains with an increase of 4.70%, while other segments like publishing and digital media saw declines [12]. - Significant developments in AI applications and gaming are highlighted, with Tencent's "Yuanbao" app reaching the top of the Apple App Store's free chart [2][17]. Market Performance - The media industry (Shenwan) increased by 0.21% during the specified period, while the Shanghai Composite Index decreased by 0.44% [12]. - The advertising and marketing sector led with a 4.70% increase, while publishing, digital media, gaming, and television broadcasting experienced declines of -0.75%, -1.31%, -1.94%, and -3.66% respectively [12]. Industry Key Data - AI Applications: Tencent's "Yuanbao" app achieved significant download growth, reaching 22.23 million downloads, with a notable 97.76% increase week-on-week [2][17]. - Gaming: The top five mobile games on iOS included "Honor of Kings" and "Peacekeeper Elite," with "Supernatural Action Group" surpassing 10 million daily active users [3][19]. - Film: The total box office for domestic films reached 283 million yuan, with "Zootopia 2" leading the weekly box office [28][30]. Investment Recommendations - The report emphasizes a positive outlook on AI applications and cultural exports, recommending focus on gaming, IP, short dramas, marketing, and publishing sectors [5][36]. - Specific companies highlighted for investment include Giant Network, Kuaishou, and Mango TV, among others [5][36].

另有AI芯片产业链人士也对《科创板日报》记者表示，未来，推理请求量与并发数将大幅增加，推理算力需求呈指数级攀升。 推理算力需求呈指数级攀升 "现阶段正处于以智能体为代表的人工智能新应用爆发初期，未来推理请求量与并发数大幅增加，推理算力需求呈指数级攀升。"一名AI产业链人士向《科 创板日报》记者表示，"随着推理范式变化，AI智能体将加速落地，其整体算力消耗可达同参数规模大语言模型的10倍以上，对智能算力的需求呈数量级 跃迁。 至2030年，预计AI推理在整个AI计算市场将占到80%的份额，而聚焦于极致推理的AI芯片，未来会有更强的爆发性。 随着大模型行业逐渐从大规模训练阶段走向推理落地阶段，业内分析普遍认为，2026年全球AI推理的需求将超过AI训练场景。 不久前，英伟达以200亿美元（约合1400亿元）收购一家AI推理芯片初创企业Groq的技术授权，并把Groq核心团队招入囊中，来补全推理算力拼图。 在近日的采访中，曦望董事长徐冰向《科创板日报》记者判断称，至2030年，预计AI推理在整个AI计算市场将占到80%的份额，而聚焦于极致推理的AI芯 片，未来会有更强的爆发性，并会对现有的算力系统造成冲击。 从市场 ...

Core Viewpoint - The company positions itself as an "integrated service provider and general operator for park innovation ecology," focusing on three leading industries: integrated circuits, biomedicine, and artificial intelligence [1] Group 1: Investment Strategy - The company is actively building an industrial investment matrix around key industries, enhancing its investment attributes towards chain-leading enterprises, unicorns, and gazelle companies [1] - It has successfully promoted investments in well-known companies such as ByteDance, Jiangbolong, Biren Technology, and SenseTime [1] Group 2: Investment Platforms and Focus Areas - The "Shanghai Lingang Ciyuan Venture Capital Fund" serves as the core platform for the company, which emphasizes strengthening its layout and cultivation of cutting-edge and disruptive technologies [1] - The company is increasing its investment support for key industries within the park, concentrating on forward-looking layouts for potential "hard tech" enterprises [1]

Core Viewpoint - The AI industry is shifting focus from training to inference, with inference requests becoming the primary demand for computational power as model training stabilizes [1][2] Group 1: Industry Trends - The demand for inference power is projected to reach 66% by 2026, surpassing training power, indicating a structural change in the industry [2] - The cost of inference currently accounts for 70% of AI application expenses, making it critical for AI companies to reduce inference costs to achieve profitability [2] - The emergence of intelligent agents and complex AI applications is accelerating the need for real-time interaction and high-frequency responses [2] Group 2: Company Developments - Xiwang Technology launched its new inference GPU chip, Qihang S3, and the Huanshi SC3 supernode solution during its first product release event after a strategic financing round of nearly 3 billion yuan [1][3] - The Qihang S3 chip features significant advancements, including a fivefold increase in inference performance compared to similar products, and is the first domestic GPGPU inference chip to use LPDDR6 memory [6][4] - The Huanshi SC3 solution is designed for large model inference scenarios, supporting high system utilization and stability, with a cost reduction from hundreds of millions to tens of millions for equivalent inference power [6][4] Group 3: Software and Infrastructure - Xiwang has developed a comprehensive self-researched software platform that is compatible with the CUDA ecosystem, facilitating seamless migration for users [7] - The company has achieved compatibility with over 90% of major models on the ModelScope platform, enhancing its service offerings [7] - The AI native intelligent computing platform introduced by Xiwang addresses industry pain points, including high GPU resource idleness and complex operational management [9][12] Group 4: Business Model Innovation - Xiwang's business model is structured around a "Token as a Service" approach, providing various token services tailored to different customer needs [14] - The company emphasizes the importance of power costs in large computing centers and has developed strategies to enhance energy efficiency and reduce operational costs [14] - Strategic partnerships with industry leaders aim to create a collaborative ecosystem that accelerates the deployment of extreme inference computing capabilities [16][17]

Core Viewpoint - The article discusses the transition of the AI industry from "training-driven" to "inference-driven" models, highlighting the importance of cost efficiency and system stability in the delivery of inference capabilities, particularly through the launch of the new inference GPU, Sunrise S3 [2][5]. Group 1: Product Launch and Features - Sunrise officially launched its new inference GPU, the S3, at the first Sunrise GPU Summit, marking its first public appearance after raising approximately 3 billion yuan in strategic financing [2]. - The S3 chip is designed specifically for large model inference, featuring a system-level design that enhances performance and cost-effectiveness, achieving over a 10-fold improvement in overall cost-performance ratio compared to its predecessor [5][6]. - The S3 supports precision switching from FP16 to FP4, significantly improving low-precision inference efficiency while increasing memory capacity by four times compared to the previous generation [5][6]. Group 2: Cost Reduction and Efficiency - In typical inference scenarios, the unit cost of token inference with the S3 has decreased by approximately 90% compared to the previous generation, enabling scalable deployment of AI applications [5][6]. - The overall delivery cost of the new SC3-256 ultra-node solution is controlled within the range of ten million yuan, significantly lower than similar solutions in the industry that cost over one hundred million yuan [6]. Group 3: Ecosystem and Cloud Strategy - Sunrise aims to build a collaborative inference cloud to address challenges such as resource fragmentation and operational complexity in the deployment of inference capabilities [8][9]. - The inference cloud will utilize the S3 as a foundation, pooling distributed computing resources into a unified inference power pool, allowing enterprises to access model capabilities on-demand without worrying about hardware configurations [9]. - The company has initiated a "one cent per million tokens" inference cost plan in collaboration with partners, indicating a shift towards economically feasible large model inference [9]. Group 4: Strategic Collaborations - Sunrise has signed a strategic cooperation agreement with Zhejiang University to establish a joint research center focused on advanced topics such as optical interconnect GPU architecture and AI high-precision weather forecasting [10]. - The company has also formed strategic partnerships with various enterprises to promote the application of inference capabilities across industries such as transportation, manufacturing, and healthcare [10].

马年将至，备受瞩目的央视春晚"科技含量"再创新高。继追觅系机器人企业魔法原子之后，1月25日， 成立不足三年的企业银河通用机器人也或将在央视春晚"露脸"，官宣成为2026年央视春晚"指定具身大 模型机器人"。 南都湾财社记者向银河通用方面了解在此次春晚上的具体互动方式，截至发稿暂未获回应。 除了银河通用，另一家官宣成为"战略合作伙伴"的魔法原子同样有来头。据了解，魔法原子成立于2024 年初，有着"追觅系"基因，其核心创始团队曾在小米机器人团队任职。目前，魔法原子布局了通用人形 机器人、四足机器人等多条产品线。据公开信息透露，该公司正在按上市时间表推进，意图加速在二级 市场的动作。 春晚迎来新面孔 相比于此前登陆春晚的"熟面孔"宇树科技，此次高调上春晚的几家企业在业内也备受关注。 公开资料显示，银河通用成立于2023年5月，核心团队拥有深厚的"北大系"学术背景。虽然成立时间不 足三年，但其已完成多轮融资。 记者梳理发现，该企业已累计完成四轮融资，投资方阵容豪华，囊括了美团、商汤、宁德时代 （300750）、中金资本以及央视融媒体基金等产业资本与基金。 值得注意的是，在完成去年底的一轮超3亿美元融资后，银河通用 ...

2026年的1月份，国内的AI界发生了几件大事： 1月2日，GPU公司壁仞科技登陆港交所，成为港股国产GPU第一股，并且成为港交所《上市规则》第18C章（特专科技）启用以来，募资规模最大的IPO 项目； 1月9日，MiniMax（上海稀宇科技）登陆港交所，不仅创下了全球AI公司从创立到上市的最短时间记录，而且打破了快手上市创下的科技公司超购记录， 而且还是全球第一家市值突破千亿的原生AI公司。MiniMax上市，被称为是中国大模型企业在全球舞台上的"定价"时刻； 1月14日，中国汽车工业协会发布的《2025城市NOA汽车辅助驾驶研究报告》。报告显示，Momenta城市NOA搭载量41.44万辆，占第三方供应商比例约 61.06%，处于断崖式领先，其产品不但覆盖国内主流乘用车企业，且与全球排名前10大车企中的8家开展合作。 壁仞科技，MiniMax，Momenta的共同点，是这三家AI翘楚公司的创始人,均出自商汤科技。 然而，商汤前员工风光无限的同时，被称为AI时代黄埔军校的商汤，自己却被边缘化了。 成立于2014年的商汤，曾经是以图像识别为特征的AI 1.0时代的"四小龙"之首，2021年上市前共完成12轮融 ...

Core Insights - By 2028, Shanghai's intelligent computing power is expected to reach 236 EFLOPS (FP16), driven by the integration of integrated circuits and artificial intelligence as leading industries [1][4] - Shanghai is implementing the "Artificial Intelligence +" initiative to lead the digital transformation of the manufacturing sector, with the city's intelligent computing power exceeding 120,000 P [1][2] Group 1: Intelligent Computing Power Development - As of the end of 2025, Shanghai's total intelligent computing power is projected to exceed 120 EFLOPS, representing a year-on-year growth of over 100%, positioning it among the top in the nation [2] - The intelligent computing power industry ecosystem in Shanghai is being built around the computational needs of artificial intelligence, encompassing hardware, software, servers, and applications from chip manufacturing to AI development [1][4] Group 2: Policy and Infrastructure Support - Shanghai has established a three-tiered policy support system for intelligent computing demand, which includes city-level coordination, district-level enhancements, and industry-specific support [4] - The city is set to proactively build a series of intelligent computing infrastructure and accelerate the large-scale deployment of locally produced chips, aiming to create a self-controlled intelligent computing industry [5]