完整早报音频，请点击标题下方小耳机收听 【 IPO 首日，智谱创 立发 起人内部信曝光：明确 2026 年目标，提及梁文锋】 智谱 AI 1 月 8 日上市，上市当天，清华大学计算机系教授、智谱创立发 起人兼首席科学家唐杰发布内部信，宣布很快将推出新一代模型 GLM-5 。并进一步明确了 2026 年公司的目标是 " 成为国际领跑的大模型企业 " 。唐杰 提及了 DeepSeek 出现带给自己的警醒道： " 文锋 2023 年创业的时候和我聊过，当时我并没有意识到他对 AGI 如此执着，感谢他带给我很多不一样的 思考。 " 唐杰称， " 选择对 AGI 技术的执着追求，不断探索 AGI 的上界，同时精准的未来预判成为下一步智谱需要不断改进和升华的地方。 " （新浪科 技）（ 阅读拓展： 超500亿，清华系智谱敲钟，"大模型第一股" 诞生 ） 【万科公告：因到龄退休，郁亮辞职】 1月8日，万科A发布公告称，因到龄退休，郁亮先生于2026年1月8日向万科企业股份有限公司（以下简称 "公 司"）董事会提交了书面辞职报告，申请辞去公司董事、执行副总裁职务。辞去上述职务后，郁亮先生将不再担任公司任何职务。 根据《万科企 ...

展会现场，阿里云展示了基于该套件的多项解决方案。在AI眼镜场景中，融合千问VL、百聆CosyVoice等模型，实现了同声传译、拍照翻译、多模态备忘等 一体化功能；在家庭陪伴机器人方面，则可实现异常监测、告警推送、视频定位及人机对话等综合服务。 随着多模态大模型技术逐渐成熟，硬件设备通过接入大模型提升交互体验已成为行业趋势。然而，单一基础模型往往难以兼顾低成本、低时延、功能丰富与 高质量效果等多元需求。阿里云此次发布的开发套件，致力于为硬件企业与解决方案商提供低门槛、快响应、多场景的开发支持。 【环球网科技综合报道】1月8日，在阿里云通义智能硬件展上，阿里云正式推出多模态交互开发套件。该套件集成了千问、万相、百聆三款通义基础大模 型，并预置生活休闲、工作效率等领域的多款智能体与工具，旨在为AI眼镜、学习机、陪伴玩具、智能机器人等硬件设备赋予更强的感知、理解和交互能 力。 模型层面，阿里云不仅提供通义系列模型，还针对多模态交互场景推出专用模型，支持全双工语音、视频、图文等多种交互方式。据称，其端到端语音交互 时延可降至1秒，视频交互时延不超过1.5秒。 此外，套件预置了涵盖生活、工作、娱乐、教育等场景的十余款智能 ...

Group 1 - The Robot ETF (562500) has shown strong performance, rising by 0.97% and demonstrating resilience with significant trading volume approaching 1 billion yuan and a turnover rate of 3.75% [1] - Key stocks within the ETF have experienced a surge, with leading stocks like Jingpin Special Equipment and Fengli Intelligent rising over 11%, and several others hitting the daily limit [1] - Industry dynamics are characterized by three main trends: 1) Cross-industry capital influx actively exploring new sectors, 2) Collaborative efforts across the supply chain strengthening the industry foundation, and 3) Stable orders validating the commercialization process, accelerating market applications [1] Group 2 - The Robot ETF (562500) is the only robot-themed ETF in the market with a scale exceeding 20 billion yuan, covering various segments such as humanoid robots, industrial robots, and service robots [2] - Recent adjustments to the ETF's constituent stocks have increased the humanoid robot content to nearly 70%, successfully removing underperforming stocks and incorporating quality candidates, achieving a "retain the strong, eliminate the weak" strategy [2]

据阿里云公众号消息，在阿里云通义智能硬件展上，阿里云全新发布多模态交互开发套件。该套件集成 了千问、万相、百聆三款通义基础大模型，并预置十多款生活休闲、工作效率等领域的Agent和MCP工 具，除了能听、会看，还能思考并且与物理世界交互，可应用于AI眼镜、学习机、陪伴玩具、智能机 器人等硬件设备。 该套件还接入了阿里云百炼平台生态，用户不仅可以添加其他开发者提供的MCP和Agent模板，还能通 过A2A协议兼容三方Agent，极大程度地扩展了应用的能力边界，帮助企业灵活搭建业务场景。 现场，阿里云还展示了面向智能穿戴设备、陪伴机器人、具身智能等领域的解决方案。例如，在AI眼 镜领域，基于千问VL、百聆CosyVoice等模型，阿里云打造了感知层、规划层、执行层以及长期记忆的 完整交互链路，可一站式实现同声传译、拍照翻译、多模态备忘录、录音转写功能。 此外，该套件预置十多款MCP工具和Agent，覆盖生活、工作、娱乐、教育等多个场景。例如，基于预 置的出行规划Agent，用户可直接调用路线规划、旅行攻略、吃喝玩乐探索等能力。 在芯片层面，该套件适配了30多款主流ARM、RISC-V和MIPS架构终端芯片平台，满 ...

Core Insights - The next phase of AI will focus on generating returns rather than just providing tools, emphasizing the need for AI to integrate with various business scenarios to demonstrate its value and profitability [1] - The selection of AI application scenarios is crucial for the success of corporate AI strategies, with a focus on technical maturity and value positioning [2] Group 1: AI Application Scenarios - Companies should prioritize AI applications in scenarios with high data maturity and mature AI technology that address core business pain points, optimizing costs and increasing revenues [1][2] - Effective scenario selection allows companies to concentrate resources strategically, enhancing competitiveness in the market [2] Group 2: Characteristics of Valuable Scenarios - Valuable AI scenarios should possess innovation, deep integration into business systems, and reusability to solve common industry problems with low replication costs [2] - The "Innovation Scene 50" initiative by TiMedia aims to recognize breakthrough scenarios in new technologies and fields that have demonstrated practical application value [2] Group 3: Awarded Scenarios - **Intelligent Education Scenario**: iFLYTEK's AI learning machine offers personalized learning through a closed-loop system, enhancing educational outcomes [4] - **Automotive Finance Scenario**: Yixin's AI model integrates various capabilities to automate the entire automotive finance process, improving efficiency [5] - **Human Resources Scenario**: Core International's AI recruitment model enhances the recruitment process through automation and data analysis, improving efficiency and accuracy [6] - **Multimodal Interaction Scenario**: Alibaba Cloud's development kit customizes intent recognition models for AI hardware, enhancing application capabilities [7] - **AI Marketing Scenario**: Hisense's collaboration with Amazon Cloud Technology enables rapid generation of advertising materials, creating new revenue streams [8] - **Intelligent Investment Advisory Scenario**: Jiufang Zhitu's AI model facilitates diverse investment advisory needs, achieving high user satisfaction [9] - **AI Wealth Management Scenario**: Yingmi Qieman's service standardizes financial capabilities for efficient AI integration [10] - **Retail Customer Service Scenario**: NetEase Cloud's AI model enhances customer service through dynamic understanding of customer intent [11] - **Energy Trading Scenario**: Yuanli Guangnian's AI model autonomously monitors market conditions and executes strategies [12] - **Compliance Quality Inspection Scenario**: Zhongguancun KJ's AI system improves quality inspection accuracy and compliance in the securities industry [13] - **Medical Diagnosis Scenario**: Wisdom Eye Technology's AI platform achieves high diagnostic accuracy through advanced technology integration [14] - **Intelligent Cockpit Scenario**: Mianbi Intelligent's AI assistant offers a native experience for in-car applications [15] - **Store Product Management Scenario**: Multi-point Smart's AI system significantly reduces waste and improves sales rates in fresh produce [16] - **Express Logistics Scenario**: Shenzhen Qianhai's AI system integrates multiple models for efficient logistics management [17][18] - **Smart Transportation Scenario**: Shanshu Technology's AI decision-making platform optimizes resource allocation in transportation [19]