对产业而言，明确的规则就是最好的营商环境——头部平台不必在灰色地带试探，中小企业不必担心被 无底线挤压，资本不必再押注"垄断式赢家通吃"，整个行业的资源配置会更有效率。 从产业演进角度，这份指引将深刻改变平台的竞争范式。过去平台比拼的是用户规模、商家排他、补贴 力度、流量霸权；未来比拼的将是技术能力、产品体验、生态效率、服务质量。强制商家站队、滥用流 量支配地位、靠低价倾销驱逐对手的模式难以为继，自营与第三方公平共处、跨平台互通、数据合理流 动、算法公开透明，会成为主流商业伦理。这不仅利好中小商家与创业者，也将倒逼头部平台把资源投 向硬核技术、产业互联网、真实价值创造，而非重复的流量博弈。 《指引》并不否定平台经济的规模效应与网络价值，而是制止以垄断为名的掠夺式经营。大型平台应放 弃功能堆砌与生态垄断，转向开放、合规、可持续的发展路径。平台的核心竞争力，从来不是锁住用户 与商家，而是持续创造不可替代的产业价值。 随着《互联网平台反垄断合规指引》（下称"指引"）于2月13日正式发布，中国平台经济进入规则清 晰、边界明确、监管常态化的全新阶段。 从产业视角看，它的核心价值在于：终结过去以规模、流量、排他性为核心的 ...

Core Insights - The article highlights the rapid growth of Soda Music, which has reached 120 million monthly active users, nearly matching NetEase Cloud Music, due to its low-price strategy and integration with ByteDance's ecosystem [1][3][6] Group 1: Growth Factors - Soda Music's monthly active users increased by 90.7% year-on-year, while other platforms saw declines, indicating a significant market shift [1] - A key turning point was the copyright bundling strategy implemented by ByteDance, which allowed Soda Music to access a library of 50 million songs, enhancing its content offerings [3] - The platform's recommendation algorithm is noted to be highly effective, allowing even free users to discover appealing music [3][4] Group 2: Content and User Engagement - The integration of AI-generated music and various content types, such as covers and live versions, has created a robust content supply chain that aligns with user preferences [3][4] - The launch of the "Soda AI Music Creation Lab" has provided creators with a comprehensive environment for music production, facilitating the rise of AI artists [4][6] - Soda Music's user engagement is further enhanced by its deep integration with Douyin (TikTok), creating a seamless transition from short video content to music consumption [6] Group 3: Pricing Strategy - Soda Music employs a low-cost membership model, offering free trials and low-priced subscriptions, making it attractive to users compared to competitors [7][9] - The pricing for VIP memberships is significantly lower than that of QQ Music and NetEase Cloud Music, catering to the current consumer willingness to pay [9] Group 4: Industry Context - The online music market in China is evolving, moving away from a "copyright is king" model to a more integrated approach that includes content generation and distribution [10][12] - The emergence of Soda Music has disrupted the existing market dynamics, challenging established players like Tencent Music and NetEase Cloud Music [12][14] - The article identifies three distinct development models in the online music market: copyright and ecosystem platforms, community and aesthetic platforms, and content generation and distribution platforms represented by Soda Music [16][18]

Core Insights - The article discusses the anxiety within the film industry regarding the potential impact of AI on job security, particularly with the launch of ByteDance's Seedance 2.0, which is touted as a powerful video generation model [1][2] - There is a fundamental debate between two factions in AI video generation: the "secular faction," which focuses on data-driven style imitation, and the "physical faction," which aims for a deeper understanding of physical laws and causality [4][3] Group 1: Technology and Market Dynamics - Seedance 2.0 optimizes the conversion rate from "director's intent to pixels," allowing for rapid video generation from prompts, significantly reducing production time [5][6] - However, Seedance 2.0 has structural limitations as each generated video is a one-time product that cannot be reused or interacted with, locking the secular faction into a "content consumption" model [7][8] - The physical faction, on the other hand, aims to create reusable 3D environments that can be applied across various industries, potentially tapping into a trillion-dollar market [8][12] Group 2: Competitive Landscape - The competition between ByteDance's Seedance 2.0 and Kuaishou's Keling AI is intensifying, with both companies vying for market share in video generation capabilities [15] - International players like Runway and Veo 3.1 are also iterating on control and physical simulation, further complicating the competitive landscape [16] - The long-term advantage of the physical faction lies in its ability to create reusable assets, while the secular faction may struggle to adapt to this evolving market [13][16] Group 3: Business Model and Future Outlook - Despite the technological advancements of Seedance 2.0, its core value remains at the "content consumption level," which may limit its long-term commercial viability [17][18] - ByteDance is advised to focus on B2B opportunities while maintaining a presence in the physical faction, rather than fully committing to one direction [19] - The true challenge for ByteDance lies in mastering distribution rights in the AI video era, as the foundation of future interactions will shift from screens to spatial environments [21][22]

10月初，市场分析机构 Adobe Analytics 发布报告指出，2025年秋冬购物季，AI驱动的购物模式将迎来爆发，仅AI带来的购物流量同比预计增长520%，并 将在感恩节前10天达到峰值。与2024年相比，美国零售网站的AI流量激增805%——而在去年，沃尔玛的「Sparky」、亚马逊的「Rufus」等AI购物工具尚 未出现。 更极致的变化先一步在中国发生了。千问APP在"春节30亿大免单"活动推动下，仅9小时订单便突破1000万单，6天完成1.2亿单，连续登顶下载榜。 AI超级应用引发的消费形式变革已势不可挡。 壹览商业注意到，酒旅业在供给侧的反应速度超乎预期。2月11日，阿里千问与飞猪宣布与全球40多家旅行品牌启动AI合作，为AI用户提供专属补贴与权 益。首批合作方涵盖中国东方航空、中国南方航空、汉莎航空等航司，锦江、华住、万达酒店及度假村等酒店集团，以及上海迪士尼度假区、方特主题乐 园等头部目的地与乐园品牌。 这种集体性 "抢摊"意味着，酒旅行业正将 AI 视为一个独立的、具备爆发力的战略渠道。 酒旅竞争进入深水区，AI颠覆一切？ 国内互联网竞争史，本质是一部流量入口迁移史。门户时代，入口掌握在 ...

多数数据团队都面临着同样的难题：维度建模 。这个概念其实并不复杂，但标准的解释通常都使用相同的例子，缺乏更深入 的理解。 下面是 有关 星型模式的 内容 和 常见的例子。 本文将探讨以下内容： 维度建模 最难的地方在于，大家都用同样的例子来解释它：销售表、订单表、交易表。这些例子反复出现。如果你的业务模 式与零售业不同，你可能会觉得这些例子与 你无关。 但以下方法对我有帮助：停止思考表。 如果你的公司里发生了什么事的话，想想实际发生了什么。 一位用户注册了。 一个支持工单已关闭。 传感器读数已收到。 销售表、订单表、交易表……这些通常都是电商或零售场景，与你的实际业务截然不同。如果你处理的是SaaS指标、物联网传 感器数据或医疗记录，你可能会开始怀疑这种方法是否真的适用于你。 确实如此。但你必须停止思考具体的例子，而开始思考其背后的模式。 付款已处理。 发生的事情就是 事实 。其他的一切都只是关于事情发生的时间、涉及的人员以及事件的类型等 背景信息。 拉尔夫·金博尔围绕这一区别构建了 一整套方法论。事实 是动词， 维度 是其他一切。 事实 与 维度 之间的根本区别 如何确定事实表的 粒度 每种表类型 实际包含 ...

Core Insights - The release of GLM-5 has significantly boosted the market performance of Zhipu, with a market capitalization exceeding HKD 200 billion after a more than 20% increase over two consecutive trading days [1] - GLM-5 features an expanded parameter scale from 355 billion to 744 billion, with enhanced capabilities in complex system engineering and long-range agent tasks [1][2] - The model has achieved a level of performance comparable to Claude Opus 4.5 in real programming environments, indicating that open-source models are catching up to closed-source ones [2] Model Capabilities - GLM-5 can autonomously perform complex tasks such as long-range planning and execution with minimal human intervention [1] - The model has implemented a training paradigm innovation through the "slime" asynchronous reinforcement learning infrastructure, significantly increasing training volume and enabling high-frequency, fine-grained iterations [2] - GLM-5 has achieved state-of-the-art (SOTA) performance in multiple evaluation benchmarks, including BrowseComp and MCP-Atlas [2] Pricing Adjustments - Zhipu has announced a structural adjustment to the GLM Coding Plan pricing, with an overall increase starting from 30%, effective February 12, 2026 [3] - The new pricing structure includes the cancellation of first-purchase discounts while retaining seasonal and annual subscription discounts [3] - Post-adjustment, the input price for GLM-5 will be up to CNY 6 per million tokens, and the output price will be up to CNY 22 per million tokens [3] Business Strategy - Zhipu aims to increase the proportion of API business revenue to 50%, with GLM-5's capabilities expected to accelerate growth in this area [4] - The company is transitioning from a focus on localized deployment to becoming a MaaS (Model as a Service) provider, anticipating a rise in both volume and pricing for API services [5] Competitive Landscape - Zhipu operates independently of major tech companies, which are generally perceived to have inherent advantages in AI due to their financial resources [6] - The company has managed to maintain lower operational costs by renting computing power rather than purchasing it outright, contrasting with the significant investments made by larger firms [6][7] - Zhipu's strategic partnership with Parallel Technology provides substantial computational support, crucial for the development and deployment of the GLM series [7]

Core Insights - The article discusses the shift in consumer behavior from traditional search methods to conversational AI interactions, highlighting a growing "invisible anxiety" among brands as they adapt to this change [1][2]. Group 1: Consumer Behavior Changes - Consumers are increasingly using conversational queries when seeking product recommendations, which reflects a significant change in search behavior [2][3]. - Real user inquiries are more complex and nuanced, often involving multiple variables such as demographics, health conditions, budget, and preferences [4][6]. - Over 80% of user questions are "combination punches," incorporating at least two complex variables, while AI-generated keyword expansions only cover about 50% of this complexity [5][4]. Group 2: Interaction Dynamics - Real conversations involve multiple rounds of questioning, with 45% of users engaging in two or more follow-up questions, revealing deeper needs and leading to more informed decisions [11][13]. - The nature of the questions posed significantly influences the AI's responses and the visibility of brands in search results [14][16]. Group 3: Source of Information - Different AI models exhibit varying preferences for information sources based on the specificity of user questions, with more detailed inquiries leading to references from local and commercial sources [18][20]. - The analysis of brand visibility shows that user-generated questions lead to a greater emphasis on lifestyle and purchasing guidance, while generic queries rely more on health information [18][20]. Group 4: Strategic Implications - The findings suggest a need for brands to adapt their marketing strategies to focus on human-centric approaches, leveraging the detailed insights from real user inquiries to enhance AI learning [22][23]. - Brands must ensure their marketing efforts align with the nuanced needs of consumers to effectively compete in the AI-driven landscape [22][23].

马年来了，全网都在"抓马"，力度等同于"抓钱"。最近火爆的商品也大多跟"马"有关，传统类的"马年生肖产品"依旧能有"理财"的效果。总之，现在 沾"马"就要火。 马年来了，全网都在"抓马"，力度等同于"抓钱"。 因为马在中国文化中象征力量、忠诚、进取、健康、吉祥等多种积极的品质和精神，谁都喜欢。即便见面问候一句"马上发财、马上幸福"，也能把亲朋好 友钓成翘嘴。 所以，商家对"马"这个吉祥物特别青睐，想尽办法要去贴一下，最为明显的是，"姓马"的明星被疯抢，人均至少2个代言。最近火爆的商品也大多 跟"马"有关，传统类的"马年生肖产品"依旧能有"理财"的效果。 总之，现在沾"马"就要火。 1. "马"姓明星代言接到手软，"外国马"成互联网顶流 最近几个月，"姓马"的明星特别走俏，无论在娱乐圈人设如何、咖位高低，只要"姓马"就会人均两三个代言，而且代言自己找上门。 比如马伊琍。几个月前，网友发帖："马年伊利会请马伊琍吗？"，还给品牌写好了广告词"我是马伊琍，马年喝伊利"，结果帖子火了，讨论声量越来越 大。 伊利特别"听劝"，立马请了马伊琍代言。做的"寻找马伊琍"的广告，点击量特别高，评论区全是各种赞美的声音，很难想象这 ...

2026年1月20日，极视角正式向港交所递交招股书，走到资本市场门前。 从招股书看，极视角是一家以计算机视觉为核心的企业级AI解决方案公司。过去三年，公司收入保持 高速增长：2022年至2024年，收入由1.016亿元增长至2.573亿元，两年复合年增长率达59.2%。 极视角的增长受益于中国CV 市场的整体扩张。招股书预测，国内该市场未来五年的复合增速将达到 37.7%。 为了在内卷中突围，极视角在2024 年开启了积极的业务迭代。招股书披露了一个关键变化： 2024年，公司引入了大模型解决方案。这一新业务在推出的当年即贡献了 6212.2 万元收入，占总营收 比重迅速达到 24.1%。 今天，硅基君就带你来看看这家即将上市的视觉AI公司。 两年复合增长59%，大模型收入6212万 2022 至 2024 年，极视角收入保持高速增长，由 1.016亿元提升至 2.573亿元，两年复合年增长率达 59.2%。2025年前9个月，公司收入为1.36亿元。 从招股书看，公司主营业务分为两大类：一是AI计算机视觉解决方案，包括标准、定制化及软硬件一 体化的产品；二是大模型解决方案，基于通用大模型，结合多智能体、RA ...

Core Insights - The AI industry is experiencing a significant moment with multiple major model releases concentrated in a short timeframe, creating a strong sense of urgency and competition among companies [1][2]. Group 1: Model Releases and Performance - In less than two weeks, several high-profile AI models have been released, including Claude Opus 4.6, GPT-5.3-Codex, Seedance 2.0, and GLM-5, indicating a competitive landscape with rapid advancements [2][4]. - GLM-5's price increase signifies strong demand and capability, with its queue exceeding initial expectations [4]. - Chinese models are not only dominating in quantity but are also achieving quality parity and even leading in some areas, with significant contributions from domestic companies [5][18]. Group 2: Market Dynamics and Trends - The emergence of GLM-5 and other models represents a shift in the AI landscape, where companies are beginning to compete on both product and model quality, particularly in the B2B sector [13][17]. - The competition is expected to intensify as more companies release models that challenge established players like Anthropic, potentially reshaping the market dynamics [12][13]. - The AI industry is anticipated to reach a critical turning point in 2026, with expectations of significant advancements and market changes [14]. Group 3: Financial Implications - Anthropic's annual recurring revenue (ARR) is projected to surpass OpenAI's for the first time in Q1, indicating a shift in financial performance within the industry [10]. - The ability of companies to monetize their models effectively is becoming increasingly important, with a focus on the economic value generated from their applications [12][20]. - The competitive landscape is likely to lead to a re-evaluation of value distribution within the industry, as companies adapt to new market realities [12][17].