Core Viewpoint - Manycore Tech Inc. (群核科技) is on track to become the "first global space intelligence stock" and aims to be the first among the "Hangzhou Six Little Dragons" to complete its IPO [3]. Group 1: Company Overview - Founded in 2011, Manycore Tech Inc. owns the largest space design platform "CoolJia" and its overseas version "Coohom," along with the next-generation space intelligence solution platform "SpatialVerse" [3]. - The company is focusing on space intelligence, emphasizing three core areas: space editing tools, space data, and space large models, creating a positive feedback loop [4]. Group 2: Technological Advancements - In early 2025, Manycore Tech open-sourced the spatial language model "SpatialLM," ranking among the top three on the HuggingFace model trend list alongside Deepseek V3 and Tongyi Qianwen Qwen2.5-Omni [5]. - The introduction of 3D Gaussian technology into AI space training aims to enhance the realism and physical consistency of simulation environments, bridging the gap between simulation and reality (Sim2Real) [5]. Group 3: Financial Performance - In the first half of 2025, Manycore Tech achieved revenue of 400 million yuan, with a gross margin of 82.1%, an increase of nearly 10 percentage points from 72.7% in 2022 [6]. - The company turned profitable in the first half of 2025, reporting an adjusted net profit of 17.83 million yuan and an adjusted net profit margin of 4.5% [6]. Group 4: Strategic Partnerships - Manycore Tech has established strategic partnerships with leading companies across various industries, including Zhiyuan Robotics, Galaxy General, and PICO, positioning itself in the key space intelligence sector [5].

Core Viewpoint - The research team led by Professor Dong Hao from Peking University has achieved a significant breakthrough by equipping robots with a "spatial brain," enhancing their ability to understand spatial commands in real environments [1] Group 1: Research and Development - The research outcome, named NavSpace, has been jointly completed by Peking University and the Qi Yuan team and published at the prestigious ICRA 2026 conference [1] - A new evaluation benchmark called NavSpace has been established, focusing on "spatial intelligence" for navigation testing, marking the first of its kind in the industry [1] - The SNav model has been developed to automatically generate a large number of training samples covering six spatial capabilities from existing data, effectively integrating spatial perception and reasoning into the model [1] Group 2: Practical Applications - Robots equipped with the SNav model can understand and execute complex instructions in real environments, such as "go to the sofa next to the second floor," showcasing the practical implications of this research breakthrough [1] - This advancement is expected to directly promote the integration of intelligent service robots into real-world scenarios [1]

Core Insights - The digital transformation of the financial industry is shifting from a focus on online expansion to enhancing efficiency and intelligence in offline spaces, as indicated by the recent policy from the National Financial Regulatory Administration [1] - The collaboration between Postal Savings Bank of China Beijing Branch and MAXHUB exemplifies a move from "tool empowerment" to "spatial intelligence" in the financial sector [1] Group 1: Challenges in Digital Transformation - Traditional financial institutions face three common challenges: inefficient internal collaboration, isolated departmental systems creating data barriers, and reliance on manual processes leading to lengthy decision chains [2] - Cross-border communication is complicated by language and cultural differences, resulting in distorted information and difficulties in establishing trust, which affects efficiency and security [2] - Customer experience remains fragmented, with services primarily focused on one-way information delivery, lacking immersive experiences and deep brand interactions [2] Group 2: Solutions Provided by MAXHUB - MAXHUB has developed a "smart collaborative network" tailored for Postal Savings Bank of China Beijing Branch, focusing on enhancing global operational efficiency, strengthening core risk control, and reshaping high-end customer connections [4] - The AI meeting solution from MAXHUB integrates real-time bilingual translation and intelligent meeting minutes into cross-border investment decision-making processes, significantly reducing communication costs and ensuring data security [5] - MAXHUB's remote collaboration and data sharing capabilities have improved risk control by connecting the Beijing Branch with the head office, enabling real-time asset quality monitoring and transforming risk management from reactive to proactive [7] Group 3: Enhancing Customer Experience - MAXHUB has created a smart digital exhibition hall for Postal Savings Bank of China Beijing Branch, transforming complex financial products into interactive and customizable visual narratives, thereby enhancing customer engagement and brand loyalty [9] - The immersive digital interactions provided by MAXHUB help deepen customer trust and understanding of the bank's offerings, contributing to a stronger brand perception [9] Group 4: Future Directions - The current practices indicate that the future competitiveness of the financial industry will stem from the deep integration of business processes and digital spaces, with offline scenarios becoming intelligent carriers of policy guidance and service delivery [11] - MAXHUB aims to continue advancing "spatial intelligence" applications in the financial sector, assisting institutions in breaking down online and offline boundaries to achieve higher quality digital development [11]

VAST是2023年3月才成立的年轻企业，自2024年初起持续迭代Tripo大模型，先后推出Tripo1.0至 Tripo3.0等一系列3D大模型，发布了TripoSR、TripoSG、TripoSF等开源3D基础模型，并配套开发了系 列DCC（数字内容创作）软件生态插件。 据央视报道，2月9日，习近平总书记来到位于北京亦庄的国家信创园，开启今年首次国内考察，并同科 研人员和科技企业负责人代表交流。 作为现场科技企业负责人代表之一，VAST创始 兼 席执 官宋亚宸在接受新京报贝壳财经专访时表示， 自己感受到了国家对未来产业和前沿创新的高度重视，也更加坚定了"接过历史接力棒"的决心。 "未来产业是抢占科技制高点、把握发展主动权的关键。"宋亚宸认为，北京的政策红利与"耐心资本"支 持，强大的原始创新能力和密集的人才储备，丰富的概念验证与应用落地场景三大优势深度融合，可以 成为北京加快建设国际科创中心的坚实路径。 他眼中，科技自立自强意味着核心技术必须自主可控，以及创新成果必须赋能产业。在核心技术上 ， VAST押注3D基础大模型，希望通过AI技术重构3D内容创作流程，构建全民化3D内容平台。 "3D抖音"是宋亚宸 ...

Core Insights - AI is described as a civilization-level technology that should involve the entire global population, not just a select few [4][5][6] - The rapid development of AI has exceeded many expectations, creating a sense of responsibility to ensure diverse perspectives are included in its evolution [6][7] - The importance of localizing AI to understand different languages and cultures is emphasized, as it affects who is included or excluded from AI systems [9][10] Group 1: AI's Impact and Responsibility - AI is a new generation of computing that will eventually be integrated into any device or system that relies on chips [5][8] - The societal implications of AI are vast, affecting various sectors such as healthcare, agriculture, and education [7][8] - The need for diverse voices in AI development is crucial, as it shapes how the technology will influence the real world [7][9] Group 2: Future of AI and Spatial Intelligence - Spatial perceptual intelligence is identified as a key frontier for AI, moving beyond language models to immersive, three-dimensional interactions [11][12] - World Labs aims to develop next-generation models that enable AI to reason, understand, and interact in three-dimensional spaces, impacting various applications [12][13] - The evolution of AI will allow for more active and agentic capabilities, enhancing its ability to model real-world interactions [13][14] Group 3: Market Applications and Innovations - The gaming industry is highlighted as a significant market for AI innovations, with World Labs already engaging developers to enhance creativity and innovation [14][15] - The tools developed by World Labs are currently being utilized by game developers, indicating a positive reception and potential for future growth in this sector [14][15]

从英伟达离职，回到中国创业，又反向杀入黄仁勋的腹地，没有投资人不喜欢这样的故事。 典型代表是摩尔线程创始人张建中，他曾任英伟达全球副总裁、大中华区总经理，在2020年创立摩尔线程，也被称作"中国版英伟达"。2025年12月，摩尔 线程登陆科创板，市值一度超过3000亿元。 另一个代表是群核科技创始人黄晓煌。在美国攻读硕士研究生期间，黄晓煌曾加入英伟达任软件工程师，参与了CUDA开发。2011年，他从硅谷回到杭 州，与同学创立群核科技，其打造的"酷家乐"是中国最大的空间设计平台。但群核科技IPO的故事，要更为曲折。 2021年4月，群核科技在纳斯达克申请上市。那时候，一起排队的中国公司有30多家。但到了下半年，风云突变，群核也无奈撤回申请。 到了2025年2月，群核科技向港交所递表，摩根大通、建银国际为联席保荐人，又正值"杭州六小龙"声名鹊起。结果6个月内，申请未获批，招股书失效。 2025年8月22日，群核再次向港交所提交上市申请，目前仍在排队中。 云启资本创始管理合伙人毛丞宇在IDG时，主导了群核的第一轮机构投资，后来又持续加注。"杭州六小龙"火了之后，有一次他跟黄晓煌感叹，"运气和 命运还是蛮奇妙的。如果 ...

近日，Behavision旗下多项面向具身智能与空间智能的核心数据资产，在北京国际大数据交易所完成登记。其中，3D铰接数据集作为全 国首个完成数据资产登记的该类数据集，具有里程碑意义。 这不仅标志着Behavision在数据要素资产化路径上实践取得关键成果，也标 志着具身智能从实验室内的数据积累，正式迈向标准化、资产化的价值流转新阶段。 近日，Behavision旗下多项面向具身智能与空间智能的核心数据资产，在北京国际大数据交易所完成登记。其中，3D铰接数据集作为全 国首个完成数据资产登记的该类数据集，具有里程碑意义。 这不仅标志着Behavision在数据要素资产化路径上实践取得关键成果，也标 志着具身智能从实验室内的数据积累，正式迈向标准化、资产化的价值流转新阶段。 本次完成登记的资产涵盖4项3D铰接数据资产与1项VLA多模态数据资产。与传统的静态3D数据不同，3D铰接数据是具身智能走向物理 世界的"说明书"，它不仅包含物体的几何形状，更精准刻画了电器、家具、日用品等真实对象的运动逻辑、关节约束及功能属性。通过 实现仿真环境与现实物理规律的深度对齐，从而为缓解具身智能领域长期存在的Sim2Real Gap ...

Core Viewpoint - ChaoTu Software (300036.SZ) is focusing on the development and innovation of AI GIS technology since 2019, with plans to launch a geospatial AI technology platform in 2024 and industry-specific intelligent systems in 2025, leading to significant growth in AI business and rapid increase in contract amounts [1] Group 1 - ChaoTu Software is recognized as a core player in spatial intelligence [1] - The company has committed to AI GIS technology research and innovation since 2019 [1] - A major breakthrough in flagship projects is anticipated, along with a substantial increase in contract amounts [1]

2026年1月28日，宸境科技在杭州举办了一场备受瞩目的新品发布会。在大会上，没有出现令大众惊叹 的机器人"炫技"表演，没有花哨的后空翻，也没有精密的穿针引线。相反，这场发布会更像是一次具身 智能"基础设施"的亮相仪式。 宸境科技跳出了单一功能的展示，向行业呈上了一套经过工业级验证的标准化感知方案，重磅发 布"LooperRobotics"品牌体系和全栈技术矩阵：Insight自主空间智能相机、TinyNav神经导航算法库以及 RoboSpatial空间感知软件平台。 感知能力的上限将决定机器人的天花板。 当具身智能的融合进入深水区，感知能力的上限将决定机器人的天花板。宸境科技试图突破空间智能技 术边界，让机器人感知技术从"定制化"走向"通用化"。在这一刻，机器人的"眼"与"脑"不再是简单的零 部件，而是被定义为赋能千行百业、驱动物理世界变革的核心基座。 硬件解构，反常识的"工程美学" 在传统的摄像头和传感器设计里，有个约定俗成的习惯：水平优先，即眼睛要横着长。这听起来很合 理，因为人眼就是横向排列的，我们习惯去关注宽广的地平线，自动驾驶汽车也需要横向视野来盯着宽 阔的马路。 然而，宸境科技这次发布的Insi ...

Core Insights - World Labs, founded by AI expert Fei-Fei Li, has launched its core product, World API, which represents a significant leap from generating video content to creating interactive 3D spaces [1][2] - The platform utilizes multimodal input technology to transform text or images into digital worlds with physical rules and geometric structures, aimed at applications in robotics training, game development, and virtual reality [1][2] - The company is currently negotiating a new round of financing, with a potential valuation of $5 billion, five times its initial funding round valuation [1][11] Group 1: Importance of World API - World API differs from mainstream AI tools by generating spatial structures rather than fixed-angle video streams, allowing for exploration and interaction within the created environments [2][3] - The distinction between video models and world models is crucial; video models address "what to see," while world models address "what can be done" [3][5] - AI must understand the functionality of objects within these environments to enable interaction, emphasizing the importance of spatial relationships and physical rules [4][5] Group 2: Core Capabilities of World API - World API supports multimodal input, allowing users to input text, images, videos, or panoramic photos to reconstruct a 3D space, overcoming the limitations of single-modality inputs [6][7] - The API incorporates semantic understanding, enabling AI to comprehend natural language descriptions without converting them into precise numerical values, thus making it user-friendly for creators [8][9] - The combination of these capabilities positions World API as a foundational platform for any system requiring 3D environments, rather than just a generative AI application [9][10] Group 3: Market Potential - The primary applications for World API include game development, where it can significantly reduce the time required for 3D modeling, and robotics training, where it can create training environments from real-world videos [9][10] - The platform also caters to VR and AR needs, lowering the barrier for creators who may not be familiar with traditional modeling software [9][10] - Future technological directions include enhancing dynamics within generated environments to improve realism and usability for physical simulations and robotics training [10] Group 4: Competitive Landscape - World Labs differentiates itself from competitors like OpenAI and Google by focusing on programmable spatial infrastructure rather than just interactive environments for specific applications [10][11] - While OpenAI emphasizes conversational agents and Google focuses on specific applications, World Labs aims to provide a versatile platform that can be integrated into various systems requiring 3D environments [11] Group 5: Ethical Considerations - Fei-Fei Li acknowledges the potential risks associated with technologies that can reconstruct spaces, emphasizing the need for responsible usage governed by laws, social norms, and education [12][13] - The company recognizes the challenges in controlling how generated spaces are utilized, highlighting the importance of ethical considerations in technology deployment [13][14] Conclusion - World API is not just a product but a gateway to a new era of spatial intelligence, allowing AI to engage with environments rather than merely generating content [15][16] - The platform is now live, and the focus shifts to how developers will utilize it to create new interactive worlds [17][18]