商业航天2.0时代，最大的对手还是美国 12月12日，长征12号如期发射。这个直径3.8米，高62米的大块头，起飞质量约430吨，近地轨道一次能运12吨，发动机和未来登月火箭用的同一款，推力 可以精细调节。为何临近年底，中国火箭发射频繁，其背后又有什么战略考量，今天有必要逐一说明。 AI不是压垮地面算力的最后一根稻草，而是把算力推上太空的第一根火柴。 当模型规模从百亿参数走向万亿参数，全球数据中心都在疯狂加电，但仍然供不应求。此时，一个被忽视多年的方向突然爆火：天基算力。 在700–800公里之外，卫星可以获得不间断太阳能供电、极低冷却成本和跨大洲低延迟覆盖，而这些特性，恰好完美适配下一代算力需求。一旦天基算力 规模化，地球将第一次拥有"无限电力+近零冷却成本"的数据中心。 这不是技术幻想，而是产业竞争的新起点。在可预期的未来，商业航天不再是火箭行业，而是"太空版信息产业革命"，至此，商业航天已经从机械重复发 射导弹的1.0时代，正式进入"太空基建"为主的2.0时代。 01 目前火箭发射频率，无法应对商业航天的2.0时代 商业航天产业链分为上游制造、中游发射、下游应用与运营，商业航天时代来临之前，上游制造成本 ...

据知情人士透露，OpenAI首席执行官山姆·奥特曼正在悄悄筹划一笔可能颠覆太空产业格局的交易，计 划通过收购或深度合作控股一家商业火箭公司，与埃隆·马斯克的SpaceX展开直接竞争。 知情人士表示，今年夏天，奥特曼与斯托克空间公司进行了首次接触，双方在秋季深入探讨了股权合作 的具体细节。 其中一个核心方案是，由OpenAI通过多轮股权注资最终获得对该公司的控股权，预计整体投资规模将 攀升至数十亿美元量级。不过，相关谈判目前处于暂时搁置状态。 算力之争，为何要上太空？ 推动奥特曼将目光投向太空的背后，是AI对算力需求呈指数级增长的冷酷现实。 他曾在公开访谈中阐述，未来的计算需求很可能将超越地球能源供给与环境承载能力的极限。在他看 来，部署于近地轨道的太空数据中心能够依托近乎无限的太阳能持续供电，且无需挤占日益紧张的地面 资源与能源网络，因而成为更具可持续性的长远解决方案。 不仅奥特曼持有这一愿景，包括亚马逊创始人杰夫·贝索斯、谷歌首席执行官桑达尔·皮查伊在内的多位 科技领袖，都曾公开表达对在太空部署计算集群构想的支持。 行动已然开始，谷歌已与卫星运营商行星实验室（Planet Labs）达成合作，计划在2027 ...

朱雀三号火箭作为我国第一发大型可回收火箭，其首发结果超出预期。二级火 箭成功进入预定轨道，完成了 ERG 分离、发动机点火及二次启动等步骤。然而， 一级火箭在着陆时出现点火异常，未能实现软着陆。尽管如此，从产业链和监 管角度来看，只要二级火箭成功入轨就算圆满完成任务。 这一成功对商业航天 产业链尤其是卫星发射有显著带动作用。目前商业航天面临严重的运力瓶颈， 大量卫星造好却无法发射。朱雀三号的成功入轨意味着明年（2026 年）民营 大型火箭公司将在发射任务中占据一席之地。预计明年液体大型火箭的发射数 量将比今年增加 30-40 次，总数达到 100 次以上，相比今年约 70 次的数量， 有显著增长。这将大幅提升卫星发射能力，预计增加数百颗卫星的运力。 此外， 这一成功对民营火箭公司的上市进展和监管审批也有积极影响。例如蓝箭航天 目前已进入券商辅导阶段，预计明年报材料。如果后续能继续取得成功，其上 首发民营大型火箭成功入轨意味着什么？20251203 摘要 朱雀三号成功入轨预示 2026 年民营大型火箭公司将在卫星发射市场占 据重要地位，预计明年液体大型火箭发射数量将显著增加，或将达到 100 次以上，大幅提升卫 ...

Group 1 - The core viewpoint is that China's space computing constellation is accelerating its construction, with the "Three-Body Computing Constellation" officially entering the networking phase, aiming to complete the deployment of over 50 computing satellites by 2025 to build an integrated space-ground computing network [1] - NVIDIA's H100 has achieved space compatibility, with computing power reaching 100 times that of previous space computers, and future tests will focus on AI processing applications [1] - SpaceX plans to accelerate its space computing layout, with the "Starship" expected to achieve an annual deployment of 1 terawatt of AI computing power, and the capacity of V3 satellites is projected to increase tenfold to 1 Tbps [1] Group 2 - The software ETF (515230) tracks the software index (H30202), which selects listed companies involved in application software, system software development, and related IT services to reflect the overall performance of the software and IT services industry [1] - This index highlights technological innovation and growth characteristics, serving as an important indicator of the development status of the domestic software industry [1]

Core Viewpoint - The launch of 12 space computing satellites marks the beginning of a new era in global space computing, initiated by the "Star Computing" program led by Guoxing Aerospace, aiming to establish a vast network of 2800 satellites in space for advanced computing capabilities [2][4][14]. Group 1: Launch and Capabilities - The 12 satellites are equipped with space computing and interconnectivity capabilities, forming the world's first space computing constellation [2][4]. - Each satellite's computing power has been enhanced from Tera-scale to Peta-scale, achieving a total in-orbit computing capacity of 5 Peta Operations Per Second (POPS) [8]. - The satellites utilize laser communication technology, enabling inter-satellite communication speeds of up to 100 Gbps [9]. Group 2: Energy Efficiency and Cost Reduction - Deploying computing power in space is an effective way to save energy costs, as satellites can harness solar energy without atmospheric interference [22]. - The cost of satellite launches is decreasing, with estimates for Chinese satellite launch costs expected to be around 60,000 yuan per kilogram in 2023 [27]. - The cold environment of space serves as a natural cooling source, reducing the energy required for cooling computing equipment [25]. Group 3: Applications and Future Developments - The space computing satellites will support real-time processing of vast amounts of deep space exploration data, enhancing scientific research capabilities [12][13]. - The "Star Computing" program plans further launches, with the next constellation already in the design phase [15][14]. - The satellites are designed to process data on-site, improving the timeliness of data applications, especially in natural disaster monitoring [28][29]. Group 4: Global Trends and Comparisons - The concept of deploying data centers in space is gaining traction globally, with initiatives in the EU and the US exploring similar projects [35][40]. - The EU has already invested 2 million euros to validate the feasibility of space data centers, while companies in the US are planning to establish data centers on the Moon [36][38]. - China's first batch of computing satellites is already operational, positioning the country at the forefront of this emerging trend [42].