Core Viewpoint - The establishment of the Star Ring OS community aims to foster collaboration among key players in the automotive and technology sectors, promoting an open-source platform that addresses industry challenges and enhances AI integration [1][3]. Group 1: Community Formation and Partnerships - The first advisory committee meeting for Star Ring OS was held, with 16 ecosystem partners from various sectors signing the community charter, including automotive chips, software services, hardware, and vehicle manufacturing [1]. - Since its open-source announcement, the company has formed partnerships with over 30 enterprises and communities [1]. Group 2: Governance Principles and AI Integration - The advisory committee established four core governance principles: openness, neutrality, sharing, and co-construction [3]. - Star Ring OS aims to integrate deeply with AI technologies, focusing on hardware-software collaboration and shared computing power as foundational elements for enhancing system capabilities [3]. Group 3: Safety and Development Standards - To ensure high reliability and safety, Star Ring OS will advance functional safety certification and establish a technical working group to create a complete safety evidence chain [4]. - The platform will provide a unified system development standard and long-term maintenance mechanisms to ensure version compatibility and consistency [4]. Group 4: Community Governance Structure - The Star Ring OS community will adopt a governance model of "Advisory Committee + Special Interest Groups (SIGs)," focusing on six key areas: hardware ecosystem and certification, AI frameworks and empowerment, core systems, functional safety, information security, and toolchains [5]. - The average issue resolution time within the community has significantly decreased from 50 days to 10 days [5].

Core Insights - The company is focusing on user preferences for vehicle features, particularly regarding the i6 model's second-row screen options, which have seen significant uptake among customers [1][2] - The i8 model's second-row television option has an installation rate exceeding 80%, while the i6's rate is over 50%, indicating strong consumer interest in these features [1][2] Group 1 - The i6 model's second-row television option has a selection rate of over 40%, with other features like premium sound and all-wheel drive also showing notable selection rates [2] - The design decisions for the L9 model, including the second-row television, were influenced by the company's founder, who initially had reservations about certain features but ultimately allowed them based on team input [2] - The company emphasizes the importance of user experience, as seen in the discussions around the i6's sunroof and structural integrity, which are designed to enhance customer satisfaction [1] Group 2 - The company is actively engaging with customers through social media platforms to discuss the operational performance and long-term fundamentals of the brand [3]

视频版： 压缩版： 理想VLA做L两个原因，技术原因是图语言的长推理能力，需要语言的token输入输出是次要的。非技术原因是更容易价值观对齐。 认为最后5%10%corner case很难靠数据或世界模型自己撞出来，而需要具备类似人的推理能力。 和业界一样，在思考GPGPU是不是AI时代的终极答案。从CPU到GPU到GPGPU，本质上是冯诺依曼架构，冯诺依曼架构核心本质是程序主要关注的是 计算不是数据，数据是第二等公民，计算是一等公民。 在AI时代，计算的算子没那么多，提出的问题是，能不能让程序更多关注数据，而不是关注计算。 理想自研的车端计算架构主要是NPU,不是SOC。SOC无非是前处理后处理的CPU Cluster，加一些IO在外面与内存访存控制器。NPU里面是一个重合架 构，加一个CCB(Central Control Computing Block)用来做一些前处理后处理，不适合非张量的计算，每个class是同构的，用Mesh Bus连在一起，也提供 Ring Bus(环形总线)做广播。原话"这个是我们完全是我们独创的一个AI推理架构，目前国内没有这么做的。" 比较挑战的是编译器（涉及很多编程模型和 ...

该群友为i8车主： 拿车有一周时间，同时也是其他新势力车企车主。趁着没有上牌，可以在各种路上放飞自我，真是越 用越惊喜，越用越喜欢，我分三方面讲讲最近的使用体验。 1.底盘试驾： 找了我们城市的奥体中心，整体路线为大环线，并且有小的上下坡，时间选在在早上，车流量不算 多，激烈驾驶可以充分检验车的驾驶乐趣。悬架分别选择低和标准，其他选择项基本上在中间位置。 总体上悬架选择低会比标准位的驾驶操控感要强（底盘低的好处），转弯同时还要加速，速度加大到 接近100，侧倾小，同时没有摆尾（其他新势力大车，不敢这样开，开快完全没有信心，担心翻车， 并且摆尾明显。我的老车也根本不敢开快。），另一个，开得很快，追到前方车辆，刹车非常好，很 难想象这么重的车能够这么好的刹停下来（有懂这块的朋友指点一下，难道是四驱的原因吗）。直线 车道，刚好遇到一辆保时捷，变道超车非常稳，我们两台车比拼速度，加速太爽了，这真是一台给驾 驶者的车。 2.自动驾驶： 自己设定交通干道，同时还有转弯、并线主路，小路也多的路线。用了几次都是0接管，可能是不特 别拥堵的原因（车还是多的），但是路线是复杂的，比如十字路同时叠加高架桥，从辅路到主路，变 道超车 ...

Core Viewpoint - The i6 orders are considered to be above good, with expectations for delivery and market performance being closely monitored in the upcoming months [1]. Group 1: Order and Delivery Expectations - By October 15, 2025, the company will have insights on potential negative public sentiment regarding i6 orders, expected order volume for the year, and delivery capacity [1]. - By January 15, 2026, the company will assess the impact of potential replacement subsidies and tax reductions on i6 orders [1][2]. - By March 1, 2026, the company will evaluate customer feedback on the i6 and other models post-Spring Festival, as well as projected sales for the first half of 2026 [2]. Group 2: Product Features and Comparisons - The i6 offers a comprehensive set of features at its price point, including a refrigerator, advanced suspension, and superior cabin technology compared to competitors [3]. - The i6 has a smaller perceived difference in quality compared to the L6 and L7 models, with notable features such as electric doors and additional comfort options [4]. - The i6 is expected to sell out its production capacity by the end of October, with potential changes in pricing strategies following the initial sales period [4]. Group 3: Market Dynamics and Challenges - The faster the i6 production capacity is sold, the more optimistic the market outlook will be, although uncertainties regarding tax increases and replacement subsidies may impact short-term orders [5]. - The complexity of market dynamics makes it challenging to quantitatively analyze the impact of known data on future orders, indicating a level of unpredictability in the automotive industry during this transitional phase [5].

Core Insights - The emergence of Vision Language Action (VLA) models signifies a paradigm shift in robotics from traditional strategy-based control to general robotic technology, enabling active decision-making in complex environments [12][22] - The review categorizes VLA methods into five paradigms: autoregressive, diffusion-based, reinforcement learning, hybrid, and specialized methods, providing a comprehensive overview of their design motivations and core strategies [17][20] Summary by Categories Autoregressive Models - Autoregressive models generate action sequences as time-dependent processes, leveraging historical context and sensory inputs to produce actions step-by-step [44][46] - Key innovations include unified multimodal Transformers that tokenize various modalities, enhancing cross-task action generation [48][49] - Challenges include safety, interpretability, and alignment with human values [47][56] Diffusion-Based Models - Diffusion models frame action generation as a conditional denoising process, allowing for probabilistic action generation and modeling multimodal action distributions [59][60] - Innovations include modular optimization and dynamic adaptive reasoning to improve efficiency and reduce computational costs [61][62] - Limitations involve maintaining temporal consistency in dynamic environments and high computational resource demands [5][60] Reinforcement Learning Models - Reinforcement learning models integrate VLMs with reinforcement learning to generate context-aware actions in interactive environments [6] - Innovations focus on reward function design and safety alignment mechanisms to prevent high-risk behaviors while maintaining task performance [6][7] - Challenges include the complexity of reward engineering and the high computational costs associated with scaling to high-dimensional real-world environments [6][9] Hybrid and Specialized Methods - Hybrid methods combine different paradigms to leverage the strengths of each, such as using diffusion for smooth trajectory generation while retaining autoregressive reasoning capabilities [7] - Specialized methods adapt VLA frameworks to specific domains like autonomous driving and humanoid robot control, enhancing practical applications [7][8] - The focus is on efficiency, safety, and human-robot collaboration in real-time inference and interactive learning [7][8] Data and Simulation Support - The development of VLA models heavily relies on high-quality datasets and simulation platforms to address data scarcity and testing risks [8][34] - Real-world datasets like Open X-Embodiment and simulation tools such as MuJoCo and CARLA are crucial for training and evaluating VLA models [8][36] - Challenges include high annotation costs and insufficient coverage of rare scenarios, which limit the generalization capabilities of VLA models [8][35] Future Opportunities - The integration of world models and cross-modal unification aims to evolve VLA into a comprehensive framework for environment modeling, reasoning, and interaction [10] - Causal reasoning and real interaction models are expected to overcome limitations of "pseudo-interaction" [10] - Establishing standardized frameworks for risk assessment and accountability will transition VLA from experimental tools to trusted partners in society [10]

2025年9月25日雷军年度演讲表示"我觉得 Modely真是一辆好车，如果不选 YU7，可以选择 ModelY。当然，国内别的车也不错，明天晚上理想i6发布， 也可以考虑一下理想i6。" 2024年4月27日雷军抖音视频表示，如果这条视频过50万点赞，将买一辆理想L6。 2024年4月28日李想表示： 2024年4月28日雷军表示： 加微信，进群深度交流理想实际经营情况与长期基本面。不是车友群。 ...

视频版： 理想i6立项的时候，我跟他说我们要做一个这个世界上从来没有出现过的全新产品，2003年初我买了 人生中的第一辆车，大众POLO 1.4。21岁的我开着它不知天高地厚的和切诺基普拉多一起爬楼找楼 去承德坝上，那台车开了10多万公里，大大的拓展了我所能触及的世界。 文字版： 2005年我创办了汽车之家，我又买了一辆宝马325i，我经常开着它去金港的赛道日，去怀柔密云跑 山。掌握方向盘，就是掌控自己的命运，那种感觉和创业是一样的。后来我又买了很多车，我也造了 接近150万辆车，理想i6要成为最好的电动车，最好的人工智能终端，它既满足年轻人的需求，也必 须让40多岁的我可以重返20岁，年轻就应该掌控命运，与众不同。 补充阅读：《 32岁李想: 这十年我买过的车 》 加微信，进群深度交流理想实际经营情况与长期基本面。不是车友群。 ...

Core Viewpoint - The article emphasizes that advanced driver assistance systems (ADAS) should be provided without additional usage fees, as exemplified by the Li Auto i6, which will come standard with AD Max and have no usage fees, allowing more consumers to experience the benefits of artificial intelligence in vehicles [1][2]. Group 1: Electric Vehicle Market - The first half of the electric vehicle market focused on electrification, which has now been widely adopted, providing consumers with benefits such as quieter operation, environmental friendliness, lower costs, faster acceleration, and reduced maintenance costs [1]. - The key benefit of smart technology currently lies in advanced driver assistance, with future potential in autonomous driving. However, many current ADAS require additional fees, limiting consumer experience with the latest smart technologies [1]. Group 2: AD Max System - Li Auto's AD Max system is positioned as one of the top performers in the market, with the company asserting that it is either first or second in the rankings, with ongoing improvements expected to enhance its recognition [2]. Group 3: Pricing and Business Models - Discussions around pricing for AI agents suggest that costs should be significantly lower than hiring a human driver, with estimates ranging from 2,000 to 3,000 yuan per month compared to 10,000 yuan for a human driver. This could lead to a more cost-effective solution when considering insurance and charging costs [3]. - The debate on whether to charge for advanced driver assistance hinges on whether it is considered a basic need for smart vehicles. Some argue that if it becomes a productivity tool, a subscription model could be justified [4][5]. Group 4: Competitive Landscape - The establishment of a new product's business model is likely to follow the formula of user value versus payment value, considering the incremental value of new experiences over old ones and the costs associated with switching [5]. - The discussion also highlights that subscription models often arise from service providers having marginal costs, while on-device models do not incur such costs, suggesting that charging for on-device AI features may not be necessary [7].

https://weibo.com/3957056582/Q5L5g9ZH3 理想 i6 宣传片来了，片子里有很多场景、也有很多情绪，最重要的是敲击前备箱两下，前备 箱就打开了，这个巨大的核心技术，理想也掌握了。 还有就是宣传片里面有很多信息，比如举个例子： 1）宣传片的主角不再有小朋友，主要是年轻群体，主角大概是25-35 岁、具备独立思考能力 的年轻群体，包括新职场人、都市青年、新锐创业者等。一开始的你应该这样。你应该那样。你不应 该这样。你不应该这样，你不应该那样。 i6官方宣传片： 原作者：微博用户马生角陈宾利 链接： 2）当用户 "想停下就停下" 时，理想 i6 是 "躺平、留白的空间"，适配松弛的生活节奏；这就 是理想i6大空间的优势。 3）无论是 "淋雨""划水" 还是 "冒险"，理想i6能成为用户践行自我生活方式的 "伙伴"，而非单 纯的代步工具。 总结下，就是9月26日的理想i6发布会，理想i6发布会应该有点东西。人生有一千种颜色，不 要放弃自己的色彩，理想i6年轻群体的有一款好产品，值得期待下。 理想第三产品线产品营销负责人若愚转发了这条微博并评论"有人看懂了 " 这就是人群面临 "社会规 ...