专题：雷军2025年度演讲：谈造芯坎坷、造车争论、赞ModelY、抢苹果用户 新浪科技讯 9月25日晚间消息，在今日的2025雷军年度演讲活动结束后，小米集团董事长、CEO雷军接 受了媒体采访。 雷军谈到过去五年小米经历了艰难转身，"大家的很多批评，我们自己也无比焦虑，所以在一个内外的 压力下，开始重新思考何去何从，并进行改变。" 他表示，"我们充分意识到要从互联网公司坚定走向智能制造的新战场，要转型成硬核的科技公司。互 联网公司是机会驱动的，科技公司是使命驱动的。所以我们坚持技术立业，一上来就制定了5年投1000 亿的计划。" 他谈到，2019年小米全年的研发费用75亿，这个情况下要投1000亿是很大的数字，"但是不管风吹雨 打，我们坚定的完成了这笔投资，大概投了1020亿~1050亿。我们新5年的规划是2000亿的研发投入。 所以大家今天看到的变化，其实这背后就有海量研发投入的结果。" 雷军进一步表示，"很多东西如果你没有足够的研发投入，你是不可能做到。我相信这次大家从小米17 看到了无数的领先技术。以前我们全是在跟随，而今天讲的绝大部分技术都是领先时代的。所以我觉得 这是小米从互联网公司坚定的走向硬核科 ...

（文章来源：新京报） 新京报贝壳财经讯（记者陈维城）9月25日晚间，在2025雷军年度演讲上，小米集团创始人、董事长兼 CEO雷军介绍，小米YU7交付的最新进展：上市3个月，交付已超过40000辆。 ...

Core Insights - Huawei and SAIC have launched the first model of the Shangjie brand, the Shangjie H5, priced below 200,000 yuan, marking a significant entry into the family car market [2][3][4] - The HarmonyOS 5 ecosystem has shown remarkable growth, with over 17 million devices connected as of September 20, 2023, indicating strong adoption across various product categories [2][7][8] Product Launch - The Shangjie H5 is priced starting at 159,800 yuan, with the high-end version featuring Huawei's ADS 4 priced at 179,800 yuan, breaking the previous price barrier of 200,000 yuan for Huawei's smart driving family [3][4] - The vehicle received over 160,000 pre-orders, with 10,000 orders placed within the first hour of its launch [4][5] Market Strategy - Huawei's executive Yu Chengdong highlighted the challenges of producing low-cost vehicles, stating that the company has historically struggled with profitability for cars priced below 300,000 yuan [3][5] - The collaboration with SAIC is credited for achieving cost control and supply chain advantages, enabling the competitive pricing of the Shangjie H5 [3][5] Ecosystem Development - Despite slower progress in the automotive sector, the overall HarmonyOS ecosystem is rapidly expanding, with a significant increase in device adoption from 10 million to 17 million in just over 50 days [7][8] - The HarmonyOS 5 version has been designed to operate independently of Unix and Linux, enhancing its appeal and functionality [7][8] Future Prospects - Huawei aims to develop a robust developer ecosystem, with over 30,000 HarmonyOS applications currently in development, targeting a goal of 100,000 applications by 2025 [8][9] - The company has initiated the "Tiangong Plan," committing 1 billion yuan to support AI ecosystem innovation, indicating a strategic focus on enhancing the HarmonyOS platform [9][10]

【雷军：小米YU7上市3个月，交付超4万辆】 #小米YU7上市3个月交付超4万辆#】 新京报贝壳财经讯（记者陈维城）9月25日晚间，在2025雷军年度演讲上，小米集团创始人、董事长兼 CEO雷军介绍，小米YU7交付的最新进展：上市3个月，交付已超过40000辆。 ...

Core Viewpoint - Xiaomi Group's CEO Lei Jun held his sixth annual speech, highlighting the successful delivery and performance of the YU7 model, which has delivered over 40,000 units since July 6, 2023 [3]. Group 1: Product Performance - The YU7 model has achieved a real-world endurance of 784 kilometers based on multiple tests, with a highway test conducted by Dongche Di showing a range of 652 kilometers [3].

新浪声明：所有会议实录均为现场速记整理，未经演讲者审阅，新浪网登载此文出于传递更多信息之目 的，并不意味着赞同其观点或证实其描述。 责任编辑：梁斌 SF055 雷军说，今年的4月1号 Su 7 Ultra量产车正式冲击了纽北，这一次天气很好，小米居然只用了4圈就留下 了两项震撼人心的成绩。 "回顾这三年的历程，经历了太多的挫折，但最后的结果远远超出了大家的想象"，雷军说，"网上有句 话，我非常喜欢，今天也特意分享给大家——你只管努力，其他的交给天意"。 专题：雷军2025年度演讲 新浪科技讯 9月25日晚间消息，在今日的2025雷军年度演讲活动上，小米集团董事长、CEO雷军发表演 讲。 ...

Core Insights - Xiaomi Group's Chairman and CEO Lei Jun held his sixth annual speech themed "Change," discussing the company's evolution and the challenges faced in the automotive sector, particularly with the YU7 model [2] - Lei Jun emphasized the importance of adopting new perspectives and bold experimentation to achieve different outcomes, highlighting Xiaomi's significant achievements that have transformed public skepticism into confidence [2] Company Developments - Lei Jun shared a personal anecdote about his friend Chen Nian, founder of Vancl, who was inspired by the SU7 launch to pivot his company towards live-streaming e-commerce, demonstrating the potential for transformation in business models [2] - Chen Nian's transition to live-streaming e-commerce has proven successful, with Vancl achieving the position of third in men's fashion on Douyin this year, showcasing the effectiveness of adapting to market trends [2]

Core Insights - Qualcomm's 30-year journey in China is characterized as a "growth history in resonance with the Chinese communication industry" [2] - The company emphasizes a long-term commitment to technology innovation and collaboration with the Chinese industry chain, rather than relying on explosive growth events [2][3] - The Snapdragon Summit serves as a platform for Qualcomm to showcase its technological achievements and deepen its commitment to the Chinese market [8] Qualcomm's Development in China - Qualcomm began its deep engagement with the Chinese market in the mid-1990s, participating in the transition from analog to digital communication through CDMA technology [2] - The 3G era marked a significant milestone, with Qualcomm collaborating with all three major Chinese operators and supporting the development of TD-SCDMA technology [2] - In the 4G era, Qualcomm partnered with Chinese smartphone manufacturers to achieve global coverage for Chinese brands [2] Automotive Sector Engagement - Qualcomm has been a key player in the automotive sector for over 20 years, providing solutions that have integrated into the automotive supply chain [3] - The company has supported over 210 models from various Chinese automotive brands in the past three years, capitalizing on the rapid adoption of new technologies [3] - Qualcomm's business model focuses on a "customer success equals self-success" philosophy, fostering a symbiotic relationship with the Chinese industry chain [3] AI and New Market Opportunities - AI was a central theme at the Snapdragon Summit, with a focus on edge AI as a critical direction for Qualcomm's technology deployment [4] - The company has been proactive in the field of embodied intelligence and robotics, showcasing applications at previous events and collaborating with leading Chinese enterprises [4] - Qualcomm is committed to developing specialized chips for emerging technologies like XR and AI glasses, addressing key challenges such as size and power consumption [4] Industrial Sector Strategy - Qualcomm is addressing the fragmented nature of the Chinese industrial market through its Leap Dragon brand, offering customized products that support NFC [6] - The company has initiated several 5G connectivity projects in collaboration with partners, demonstrating the potential for smart manufacturing in China [6] - The future market for robotics and AR/VR/AI glasses is expected to be substantial, with Qualcomm aiming to drive the development of killer applications in these areas [6] Innovation and Compliance - Qualcomm's development strategy is defined by "innovation and cooperation," focusing on continuous self-challenge and enhancing collaborative capabilities [7] - The company emphasizes the importance of balancing technological innovation with compliance, particularly in the context of AI and data security [7] - Qualcomm is adapting to the trend of self-developed chips among manufacturers by strengthening partnerships and diversifying its business across various sectors [7] Market Dynamics and Future Outlook - The initial influx of players in emerging markets is seen as beneficial for validating market value and accelerating technological breakthroughs [8] - Qualcomm's historical experience suggests that market consolidation will eventually occur, with a few dominant players emerging in various sectors [8] - The company remains committed to long-term investment in technology innovation and exploring opportunities in AI and 6G alongside the Chinese industry chain [8]

Core Viewpoint - 2025 is a pivotal year for automotive intelligence, initiating a three-year cycle that will drive domestic electrification penetration rates to achieve a leap of 50%-80%+, reshaping the vehicle landscape [4][12]. Group 1: Industry Trends - The leading intelligent driving manufacturers have successfully implemented urban NOA experiences, including complex scenarios like roundabouts and U-turns, while enhancing high-level functions such as parking and scene understanding [4][12]. - The report includes both large-sample centralized road tests and small-sample in-depth road tests, evaluating the intelligent driving experiences of ten manufacturers, including XPeng, NIO, and Tesla, across various dimensions [5][12]. Group 2: Performance Improvements - Compared to Q1 2025, Q3 2025 saw improvements in intelligent driving capabilities across all manufacturers, with narrowing absolute gaps. The first-tier manufacturers achieved full-scene NOA and continued to optimize user experience [6][12]. - New entrants in self-research have shown impressive intelligent driving performance, with expectations for significant iterations in the next six months, potentially leading to substantial changes in overall experience [7][12]. Group 3: Specific Manufacturer Insights - XPeng, Huawei, and Li Auto remain in the first tier, with XPeng's self-developed Turing chip achieving 2250 TOPS of effective local computing power, enhancing driving logic [7][12]. - NIO and Xiaomi have rapidly caught up to near-first-tier levels, transitioning from "usable" to "well-usable" experiences, with NIO's world model exceeding expectations and Xiaomi enhancing complex scenario handling capabilities [8][12]. Group 4: Road Test Results - XPeng's average total takeover count was 1.51, indicating excellent overall performance, particularly in roundabout capabilities and efficiency [46][49]. - The average takeover count for the Huawei ADS-equipped vehicle was 0.60, with high scores in stability and comfort during driving [50][52]. - Li Auto's average takeover count was 1.47, demonstrating strong performance in roundabout scenarios and overall driving stability [53][56]. - NIO's average takeover count was 2.03, with good performance in roundabouts and a generally smooth driving experience [57][60]. - Xiaomi's average takeover count was 1.94, showing stable performance and the ability to handle complex scenarios effectively [60][62].

Core Viewpoint - Alibaba Group's CEO emphasizes that AI will be the next generation operating system, with aspirations towards Artificial Superintelligence (ASI) [1] Group 1: AI in Automotive Industry - The automotive industry is increasingly integrating AI agents into smart cockpits, with companies like Li Auto, BYD, and NIO leading the charge [1][2] - Li Auto's AI agent can perform tasks such as ordering food through voice commands, showcasing the practical applications of AI in vehicles [1] - The goal is to create a closed-loop system for users that connects "people, cars, and life" through the vehicle's AI capabilities [2] Group 2: Functionality of AI Agents - AI agents in vehicles primarily assist with navigation, food ordering, and ride-hailing, but require external ecosystem connectivity to function effectively [3] - Li Auto's AI agent operates on two frameworks: CUA (Cockpit Using Agent) and MCP/A2A (Multi-Channel Processing/Agent-to-Agent), allowing for task execution through third-party applications [3][5] - CUA is the prevalent technical route, enabling multi-modal understanding and task execution via apps like Alipay [3] Group 3: Challenges and Solutions - CUA faces limitations, particularly in complex tasks where accuracy is low, with only about 30% success in multi-step tasks [4] - Li Auto proposes a solution by breaking down tasks into manageable steps, improving prediction accuracy for user actions [4] Group 4: Future Developments - The vision for Li Auto's AI agent includes "full memory" and "environmental perception," allowing the agent to remember user actions and recognize environmental cues [7] - Full memory encompasses user interactions and relationships, enabling the AI to proactively assist rather than just react [7] - Environmental perception is crucial for the AI to interact with real-world data, enhancing its ability to perform tasks effectively [8]