贾鹏一字不差原话："其实是跟Tesla团队交流会比较多，我们会发现，其实Tesla的V14,它那时候已经 V13出来，然后在做V14的版本，发现跟我们思路是一模一样，我当时就感觉又惊喜又有点失望说实 话。惊喜的是它跟我们的想法是一样的，下一代用的就是一个一体化的VLA这种感觉，因为他把世 界模型跟VLA合在了一起，我们的想法也是类似。 然后第二个呢，我觉得稍微有点失望，就是他竟然比我们还晚。因为它是最近一代才开始把语言的能 力放进去，去做System 2的这种长链条的CoT，但是我们24年就已经把它量产。然后同时呢也跟NV的 团队去聊，就是Tesla进展到哪一步。然后也看它在Austin有robotaxi的进展。" TOP2备注:和特斯拉团队交流知道V14思路和理想一模一样时间线是2024年底来自前面交流的内容。 原话出处见视频53min01处 晚点2026年3月23日文章里没有没提" Tesla团队交流会比较多 "这个点，然后没提贾鹏认为V14思路和 理想一模一样这个点，用的是贾鹏后面说的特斯拉世界模型和VLA结合思路和理想类似。 2026年3月24日播出的内容里黄仁勋说 2025年10月24日，随着 特斯 ...

Core Viewpoint - The Vice President of Horizon, Lv Peng, affirmed that the company will not abandon its end-to-end team even if it launches the VLA, emphasizing that a strong end-to-end foundation is essential for the success of VLA [1] Group 1 - Horizon's commitment to maintaining its end-to-end team is seen as crucial for the development of its new VLA product [1] - Lv Peng believes that without a solid end-to-end system, the VLA would struggle to perform effectively [1]

Summary of Key Points from Conference Call Records Industry Overview - The humanoid robot solid-state battery demand is projected to exceed 74 GWh by 2035, indicating significant growth potential in the robotics and battery sectors [1] - Shandong aims to achieve a scale of over 200 billion yuan in the robotics and intelligent equipment industry by 2026, highlighting regional ambitions for industry expansion [2] Company Highlights - ZhiYuan reported a 15-fold increase in VLA edge inference performance, validated through the Spirit G2 robot, showcasing advancements in AI and robotics technology [1] - AutomationAnywhere is in discussions to acquire C3.ai, which would facilitate its entry into the public market, indicating strategic moves in the AI software sector [1] - LG Energy will supply batteries for humanoid robots to Tesla, reflecting partnerships that could enhance product offerings in the robotics market [1] Financial Performance - Xiangxin Technology has projected a net profit for the year between 168 million yuan and 200 million yuan, representing a decline of 53.26% to 44.36%, which may indicate challenges in the company's financial health [3]

Core Viewpoint - The article discusses the advancements in autonomous driving technology, particularly focusing on the new city NOA solution based on the single journey 6M architecture implemented in the Li Auto L series. It highlights the integration of various algorithms and the overall architecture of the system, emphasizing its potential for real-world applications and safety mechanisms. Group 1: Technical Architecture - The single J6M architecture achieves end-to-end processing with reinforcement learning, showcasing a sophisticated network structure that operates on 128 TOPS computing power [3][5] - The architecture utilizes Safe RL to optimize vehicle trajectory, indicating a focus on safety and reliability in autonomous driving [5] - The input data includes time-sequenced images, Lidar, SD navigation, and vehicle positioning, which are fused to create a global BEV representation [6] Group 2: Algorithmic Innovations - The article mentions the validation of algorithms like DiffusionDrive and Flow Matching by multiple companies, indicating their readiness for mass production [5] - It introduces two recommended algorithms, Diffusion Planner and Flow Planner, with the latter being an improved version developed by a team from Tsinghua University [5] - The Flow-Matching Planner is used to decode vehicle motion predictions and multi-modal vehicle trajectories, enhancing the system's predictive capabilities [7] Group 3: Future Developments - The next-generation autonomous driving model architecture combines VLA with World Model into a unified end-to-end system, suggesting a significant evolution in the technology [11] - The article notes that the VLA system is not merely a demonstration of capabilities but serves as foundational infrastructure for the scalable operation of L4 autonomous systems [12] - The CEO of the company expresses that neither VLA nor the world model will be the ultimate solution for autonomous driving technology, indicating ongoing development and exploration in the field [13]

Group 1 - The core viewpoint expressed by the CEO of Xiaopeng Motors is that the current challenges in the electric vehicle (EV) sales are temporary and the industry will recover in due time, indicating a positive outlook for the future [1] - The CEO emphasized that the worst times can also present the best opportunities, suggesting a resilient approach to the current market conditions [1] - Xiaopeng Motors is focusing on the production of its VLA, VLM, and humanoid robots, indicating a strategic shift towards innovation and new product lines to capitalize on future opportunities [1]

Core Viewpoint - The article discusses the evolution of the autonomous driving industry in 2025, highlighting the dual focus on technology proliferation and technical challenges, with traditional automakers pushing for accessibility and new players striving for technological advancements [4][5]. Group 1: Industry Trends - In 2025, traditional automakers like BYD, Geely, and Chery are leading the charge in making autonomous driving technology more accessible by integrating mid-level highway NOA features into vehicles priced over 100,000 yuan [4]. - New entrants and leading autonomous driving suppliers are focused on pushing the limits of technology, adhering to a model of annual technological iteration [4][5]. - The industry is witnessing a bifurcation, with one camp focused on accessibility and the other on technological challenges, particularly in the realm of algorithm development [4]. Group 2: Technological Advancements - The transition from "passive perception" to "active cognition" is marked by the introduction of world models, which represent a significant paradigm shift in autonomous driving technology [5][6]. - 2025 is characterized as a year of significant technological transition, with the widespread adoption of end-to-end systems and the emergence of world models and VLA (Vision-Language-Action) technologies [6][9]. - NIO is highlighted as a pioneer in the world model space, having launched its world model in 2024, transitioning from "perception-driven" to "cognition-driven" systems [5][6]. Group 3: Data Infrastructure and Chip Development - The importance of data infrastructure is emphasized, with companies like NIO benefiting from early investments in data collection and model training capabilities [7][8]. - The year 2025 is noted as a pivotal year for integrated hardware and software solutions, with companies like NIO and XPeng achieving self-developed chip integration [7][8]. - The article warns of the risks associated with outsourced chip development, contrasting it with NIO's genuine self-development efforts, which involve significant technical team investments [8]. Group 4: Regulatory and Market Dynamics - The issuance of L3 licenses is seen as a significant step towards the next phase of autonomous driving, indicating a shift from L2+ mass production to L3 and L4 capabilities [8][9]. - While traditional automakers have secured initial L3 licenses, their capabilities are questioned, suggesting that true advancements will come from new players and those with strong model capabilities [9][10]. - The ultimate value of autonomous driving technology is framed around enhancing driver convenience and significantly reducing traffic accidents, with a focus on safety as a primary goal [9].

Group 1 - The core achievement of GESON Technology in the 2025 Shenzhen Intelligent Robot Dexterous Hand Competition, winning the championship by setting a new limit for long-range delivery tasks, is supported by RoboSense's VLA model and advanced sensor systems [1][3][9] - The competition showcased the industry's leading capabilities in robot-eye coordination technology, emphasizing the commercial strength of creating industrial value through collaboration [3][9] - The event attracted 53 high-level teams from various regions, highlighting the significance of the competition in the context of the "robot mass production year" [9] Group 2 - RoboSense's recent video release demonstrated the robot's ability to perform complex tasks, indicating the integration of its core technologies aimed at flexible automation in delivery, manufacturing, and logistics [5][13] - The competition tested robots under real-world conditions, including human traffic and elevator sharing, emphasizing the challenges faced in the last 100 meters of delivery [11][13] - The success of GESON Technology illustrates the potential for RoboSense's AI robot technology to support autonomous completion of complex tasks, establishing a comprehensive technological barrier from foundational technology to application [11][13]

Core Insights - The robotics field is still in its early stages, as highlighted by Jim Fan, NVIDIA's robotics head, indicating a lack of standardized evaluation metrics and the disparity between hardware advancements and software reliability [1][8][11]. Group 1: Hardware and Software Disparity - Current advancements in robotics hardware, such as Optimus and e-Atlas, outpace software development, leading to underutilization of hardware capabilities [14][15]. - The need for extensive operational teams to manage robots is emphasized, as they do not self-repair and face frequent issues like overheating and motor failures [16][17]. - The reliability of hardware is crucial, as errors can lead to irreversible consequences, impacting the overall patience and scalability of the robotics field [18][19]. Group 2: Benchmarking Challenges - The lack of consensus on benchmarking in robotics is a significant issue, with no standardized hardware platforms or task definitions, leading to everyone claiming to achieve state-of-the-art (SOTA) results [20][21]. - The field must improve reproducibility and scientific standards to avoid treating them as secondary concerns [23]. Group 3: VLA Model Insights - The Vision-Language-Action (VLA) model is currently the dominant paradigm in robotics, but its reliance on pre-trained Vision-Language Models (VLM) presents challenges due to misalignment with physical world tasks [25][49]. - The VLA model's performance does not scale linearly with VLM parameters, as the pre-training objectives do not align with the requirements for physical interactions [26][51]. - Future VLA models should integrate physical-driven world models to enhance their ability to understand and interact with the physical environment [50]. Group 4: Data Importance - Data plays a critical role in shaping model capabilities, with the need for diverse data sources and collection methods being highlighted [31][43]. - The emergence of new hardware and data collection methods, such as Generalist and Egocentric-10K, demonstrates the growing importance of data in the robotics field [36][42]. - The current data collection strategies remain open-ended, with various approaches still being explored [43]. Group 5: Industry Trends - The robotics industry is projected to grow significantly, from $91 billion currently to $25 trillion by 2050, indicating a strong future potential [57]. - Major tech companies, excluding Microsoft and Anthropic, are increasingly investing in robotics software and hardware, reflecting the sector's attractiveness [59].

Core Viewpoint - The robotics field is still in its early stages, with significant advancements in hardware but limitations in software reliability and performance [1][12]. Group 1: Hardware and Software Dynamics - Current hardware advancements outpace software development, leading to reliability issues that hinder software iteration speed [11][14]. - Many demonstrations of robotic capabilities are often the result of selecting the best performance from numerous attempts, rather than consistent reliability [7][22]. - The need for extensive operational teams to manage robots highlights the challenges in hardware reliability, including overheating and motor failures [18][19]. Group 2: Benchmarking Challenges - The robotics sector lacks standardized benchmarks, making it difficult to assess performance consistently across different hardware platforms and tasks [21][22]. - The absence of consensus on evaluation criteria leads to a situation where every new demonstration can be considered state-of-the-art, complicating progress in the field [22][23]. Group 3: VLA Model Limitations - The Vision-Language-Action (VLA) model, currently a dominant paradigm, faces structural issues as it is primarily optimized for visual question answering rather than physical task execution [24][50]. - The performance of VLA models does not improve linearly with the increase in VLM parameters due to misalignment in pre-training objectives [26][52]. - A shift towards video world models is suggested as a more suitable pre-training target for robotics, as they inherently encode physical dynamics [27][53]. Group 4: Importance of Data - Data plays a crucial role in shaping model capabilities, and the integration of hardware and data is essential for effective robotic performance [31][32]. - Recent advancements in hardware, such as Figure03 and others, demonstrate improved motion capabilities, but challenges remain in enhancing hardware reliability [35][37]. - The Generalist model illustrates the scaling law in embodied intelligence, where larger datasets lead to better task performance [38][41]. Group 5: Future Trends and Market Potential - The robotics industry is projected to grow from $91 billion to $25 trillion by 2050, indicating significant investment potential [60]. - Major tech companies are increasingly investing in robotics software and hardware, reflecting the sector's attractiveness despite current challenges [62].

Core Insights - The new Weipai Blue Mountain Intelligent Advanced Edition is positioned as the first "six-seat plug-in hybrid SUV" equipped with the VLA large model, starting at a limited-time price of 275,800 yuan [1] - The vehicle features advanced capabilities such as voice control, CoT reasoning cards, defensive driving, and special scenario understanding, achieving an intelligent closed-loop from perception to execution [1] - The Hi4 performance version boasts a unique four-speed full-speed direct drive technology, achieving 0-100 km/h acceleration in 4.9 seconds, a low fuel consumption of 6.5L/100km, and a comprehensive range of 1,343 km [1] Group 1 - The VLA and Hi4 systems work in synergy to create a comprehensive "perception-decision-control" chain, enhancing safety and performance in various driving conditions [2] - The collaboration between VLA as the "navigator" and Hi4 as the "driver" results in a dual safety defense of "active avoidance" and "active stability," providing a superior travel experience [2] - The Coffee OS 3.4 system integrates AI services and user-friendly interactions, creating a "five good cabin" experience with features like 23 speakers and a 17.3-inch 3K entertainment screen [2] Group 2 - The vehicle includes innovative features such as AI multi-screen expansion and a motion-sickness relief display, demonstrating the practical value of technology in enhancing user comfort [2] - The "Little Wei Classmate" feature actively senses the environment and passengers, offering thoughtful collaborative services [2]