华尔街分析师Hans Mosesmann指出，液冷技术对于克服AI云端运算挑战非常关键，能为超大规模云端 服务铺路。徐明微也表示，英伟达的这一选择佐证了液冷技术是AI算力的标配。随着AI算力市场需求 的不断扩大，芯片性能和功耗水平已提升至新量级。例如，基于Blackwell架构的B200功率达到 1000W，许多数据中心单机柜密度已经达到70 - 100kW，这些高功耗输出对液冷技术提出了更高要求。 保障AI芯片的性能与运行稳定尤为关键，这就要求液冷系统更加安全可靠。 【环球网科技综合报道】4月7日，在英伟达年度GTC大会上，液冷技术成为其新品的一大亮点，引发行 业广泛关注，液冷市场"奇点"似乎正加速到来。 英伟达新一代Blackwell Ultra芯片通过"液冷 + 硅光子"的协同进化，基于5nm工艺，单颗芯片集成288GB HBM3e显存，FP4算力达15PetaFLOPS。其DGX GB300系统采用Grace Blackwell Ultra超级芯片（包含36 块Grace CPU和72块Blackwell GPU），以及机架级液冷设计，较上一代Hopper架构可提供70倍的AI性 能。此外，英伟达推出 ...

Core Insights - Liquid cooling technology has emerged as a highlight at NVIDIA's recent GTC conference, showcasing the integration of liquid cooling and silicon photonics in their new Blackwell Ultra chip, which features 288GB HBM3e memory and achieves 15 PetaFLOPS of FP4 computing power [1] - The introduction of the open-source inference framework Dynamo and the high-performance liquid cooling system indicates a significant shift towards liquid cooling in AI computing, with expectations for rapid penetration in the market [1][2] Group 1: NVIDIA's Innovations - NVIDIA's new DGX GB300 system, utilizing the Grace Blackwell Ultra super chip, offers a 70-fold increase in AI performance compared to the previous Hopper architecture [1] - The liquid cooling system can handle a power output of 120kW, enabling calculations at the scale of hundreds of quintillions [1] - Future plans include the Vera Rubin Ultra, expected to launch in the second half of 2027, with each rack potentially reaching 600kW of power [1] Group 2: Market Trends and Demand - The demand for AI computing power is expanding, necessitating higher performance and power levels from core chips, which in turn raises the requirements for liquid cooling technology [3] - Major tech companies like Microsoft, Google, Meta, and Intel are accelerating their adoption of liquid cooling solutions, indicating a broader industry trend [4] - A report from three major telecom operators anticipates that by 2025, over 50% of projects will utilize liquid cooling technology [4] Group 3: Industry Collaboration and Development - Green Cloud Technology, a leading domestic provider of liquid cooling solutions, has established a comprehensive capability from R&D to delivery, demonstrating maturity and reliability in its products [5] - The company has delivered multiple liquid cooling projects for AI firms, emphasizing the need for safety, efficiency, and standardization across the industry [6] - Collaboration among various stakeholders, including chip manufacturers and cooling solution providers, is essential for advancing liquid cooling technology and establishing unified standards [6]

Core Viewpoint - The article highlights NVIDIA's GTC 2025 event, emphasizing the shift in AI focus from training to inference, showcasing new hardware and software innovations aimed at enhancing AI capabilities and applications [1][3][30]. Group 1: Key Innovations and Products - NVIDIA introduced the Blackwell Ultra GPU series and the next-generation architecture Rubin, with plans for the Vera Rubin NLV144 platform to launch in the second half of 2026 and Rubin Ultra NV576 in the second half of 2027 [5][10]. - The Blackwell Ultra architecture significantly enhances AI performance, achieving a 1.5x improvement in AI performance compared to the previous generation, and offers a 50x increase in revenue opportunities for AI factories [8][10]. - The new CPO switch technology aims to reduce data center power consumption by 40MW and improve network transmission efficiency, laying the groundwork for future large-scale AI data centers [13][14]. Group 2: AI Inference and Software Upgrades - NVIDIA's new AI inference service software, Dynamo, is designed to maximize token revenue in AI models, achieving a 40x performance improvement over the previous Hopper generation [19][21]. - The introduction of AI agents and the Ll ama Nemo tr o n series models aims to facilitate complex inference tasks, enhancing capabilities in various applications such as automated customer service and scientific research [20][30]. Group 3: Robotics and Physical AI - NVIDIA launched the GROOT N1, the world's first open-source humanoid robot model, designed for various tasks such as material handling and packaging, indicating a significant step towards the commercialization of humanoid robots [25][30]. - The company also introduced new desktop AI supercomputers, DGX Spark and DGX Station, aimed at providing high-performance AI computing capabilities for researchers and developers [23][24]. Group 4: Market Sentiment and Future Outlook - Despite the significant technological advancements presented at GTC 2025, NVIDIA's stock price fell by 3.43% post-event, reflecting ongoing market concerns regarding AI spending and competition [28][29]. - Analysts suggest that while there are concerns about AI capital expenditure growth in 2026, the overall sentiment may improve due to the innovations showcased at the event [29][30].