Key Takeaways INTC and Exostellar aim to boost AI efficiency with Gaudi accelerators and Kubernetes-native orchestration. The solution supports hybrid infrastructures with dynamic scheduling, quota control and multi-vendor access. INTC sees Gaudi 3 scaling AI with high throughput, Ethernet interconnect and open software flexibility.Intel Corporation (INTC) has partnered with Exostellar to make enterprise-grade AI infrastructure accessible in a cost-effective manner. Intel’s partnership with this leading i ...

金磊 克雷西 发自 凹非寺 量子位 | 公众号 QbitAI 今年，AI大厂采购GPU的投入又双叒疯狂加码—— 马斯克xAI打算把自家的10万卡超算扩增10倍，Meta也计划投资100亿建设一个130万卡规模的数据中心…… GPU的数量，已经成为了互联网企业AI实力的直接代表。 的确，建设AI算力，这种堆卡模式是最简单粗暴的，但实际上， AI集群却并非是卡越多就越好用。 GPU虽然计算性能好，但是在集群化的模式下依然有很多挑战，即便强如英伟达，也面临通信瓶颈、内存碎片化、资源利用率波动等问题。 简单说就是，由于通信等原因的限制，GPU的功力没办法完全发挥出来。 所以，建设AI时代的云数据中心，不是把卡堆到机柜里就能一劳永逸，现有数据中心的不足，需要用架构的创新才能解决。 最近，华为发布了一篇60页的重磅论文，提出了他们的下一代AI数据中心架构设计构想—— Huawei CloudMatrix ，以及该构想的第一代产 品化的实现CloudMatrix384。相对于简单的"堆卡"，华为CloudMatrix给出的架构设计原则是，高带宽全对等互连和细粒度资源解耦。 这篇论文干货满满，不仅展示了CloudMatrix ...

Core Insights - CoreWeave (CRWV) and Microsoft Corporation (MSFT) are key players in the AI infrastructure market, with CRWV focusing on GPU-accelerated services and Microsoft leveraging its Azure platform [2][3] - CRWV has shown significant revenue growth driven by AI demand, while Microsoft maintains a strong position through extensive investments and partnerships [5][9] CoreWeave (CRWV) - CRWV collaborates with NVIDIA to implement GPU technologies and was among the first to deploy NVIDIA's latest clusters for AI workloads [4] - The company reported revenues of $981.6 million, exceeding estimates by 15.2% and increasing 420% year-over-year, with a projected global economic impact of AI reaching $20 trillion by 2030 [5] - CRWV has a substantial backlog of $25.9 billion, including a strategic partnership with OpenAI valued at $11.9 billion and a $4 billion expansion agreement with a major AI client [6] - The company anticipates capital expenditures (capex) between $20 billion and $23 billion for 2025 to meet rising customer demand, with interest expenses projected at $260-$300 million for the current quarter [7] - A significant risk for CRWV is its revenue concentration, with 77% of total revenues in 2024 coming from its top two customers [8] Microsoft Corporation (MSFT) - Microsoft is a dominant force in AI infrastructure, with Azure's global data center coverage expanding to over 60 regions [9] - The company invested $21.4 billion in capex in the last quarter, focusing on long-lived assets to support its AI initiatives [10] - Microsoft has a $315 billion customer backlog and is the exclusive cloud provider for OpenAI, integrating AI models into its services to enhance monetization opportunities [12] - The company projects Intelligent Cloud revenues between $28.75 billion and $29.05 billion for Q4 fiscal 2025, with Azure revenue growth expected at 34%-35% [14] Share Performance - In the past month, CRWV's stock surged by 69%, while MSFT's stock increased by 8% [17] - Current Zacks Rank indicates MSFT as a better investment option compared to CRWV, which has a lower rank [18]

综上所述，CloudMatrix384并⾮意在成为NVIDIA H100的普适性替代品，⽽是⼀款针对 特定（且⽇益重要的）AI⼯作负载进⾏深度优化的、具有⾼度创新性的专⽤系统。它的出 在性能层⾯，论⽂数据显示，CloudMatrix-Infer服务⽅案在昇腾910C上运⾏MoE模型时 ，其计算效率（以tokens/s/TFLOPS衡量）在预填充（Prell）和解码（Decode）阶段均 超越了已公开的NVIDIA H100与H800数据。这⼀成就并⾮源于单NPU在理论峰值算⼒上 的超越，⽽是华为"以系统取胜"策略的集中体现。通过PDC解耦服务架构、⼤规模专家并 ⾏（LEP）、硬件感知的融合通信算⼦（如AIV-Direct）以及精细化的INT8量化等⼀系列 软硬件协同优化，华为最⼤化了集群的有效算⼒利⽤率。 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 更多一手调研纪要和海外投行研报数据加V：shuinu9870 ...

近日，OpenAI向开发者发了一封邮件，宣布将于7月14日正式移除 GPT-4.5 Preview API。 图注：OpenAI邮件。图源网络 对于那些已经将GPT-4.5深度集成到自己产品或工作流中的开发者来说，这无异于一次震撼。他们必须在不到一个月的时间内，从OpenAI提供的近40个模 型中，重新寻找一个替代品。 为什么非关不可？ 许多人将矛头指向了高昂的计算成本。毕竟，一个性能优越、但商业上不划算的模型，在任何一家公司的账本上都不会长久。 图注：GPT模型一览 GPT-4.5 API 定价高达 75 美元 / 百万输入 tokens，150 美元 / 百万输出 tokens，几乎是 GPT-4.1 的多倍。 OpenAI官方称，这次移除计划早在4月发布GPT-4.1时就已公布。GPT-4.5从始至终都是一个"实验性"产品，其使命是为未来的模型迭代提供经验，尤其是 在创意和写作的细微之处。邮件只是按计划发送的提醒。 不够，GPT-4.5 预览版将继续作为选项，通过应用程序顶部的下拉模型选择菜单，提供给个人 ChatGPT 用户使用。 图注：用户表示GPT-4.5是最喜欢的模型之一。 最近，OpenAI公 ...

June 13, 2025 12:00 PM GMT Morgan Stanley does and seeks to do business with companies covered in Morgan Stanley Research. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of Morgan Stanley Research. Investors should consider Morgan Stanley Research as only a single factor in making their investment decision. For analyst certification and other important disclosures, refer to the Disclosure Section, located at the end of this report. Inve ...

Key Takeaways CRWV stock soared 137% in a month, beating gains from MSFT, AMZN, and the broader tech sector. Surging AI demand led to a 420% jump in Q1 revenues, while the $11.9B OpenAI deal adds further upside. CRWV guides 2025 revenue at $4.9B-$5.1B, backed by increasing demand and a $259B revenue backlog.CoreWeave, Inc. (CRWV) stock has gained 136.6% in the past month and closed last session at $149.70, jumping more than threefold from its initial opening price of $39. It has outperformed the 5.4% grow ...

In this articleNVDA"Compared to Nvidia's export-restricted chips, the performance gap between Huawei and the H20 is less than a full generation," said Dylan Patel, founder, CEO and chief analyst of SemiAnalysis. Sinology | Moment | Getty ImagesWith the U.S. restricting China from buying advanced semiconductors used in artificial intelligence development, Beijing is placing hopes on domestic alternatives such as Huawei. The task has been made more challenging by the fact that U.S. curbs not only inhibit Chin ...

今日早盘，创业板人工智能板块盘中领涨，创业板人工智能ETF国泰（159388）涨超2%。 5月28日，英伟达披露了2026财年一季度财报。根据英伟达方面提供的数据，截至2025年4月27日的2026 财年第一季度，英伟达实现收入441亿美元，较上一季度增长12%，较去年同期增长69%，其中，数据 中心同比+73%，Blackwell 芯片贡献数据中心收入的70%。黄仁勋表示，Blackwell NVL72目前正通过全 球领先的系统制造商和云服务提供商进入全面量产阶段，AI推理的token生成量在短短一年内激增十 倍。（提及个股仅为说明观点，不构成投资建议，下同） AI 基础设施服务商 CoreWeave 自上市以来持续上涨，尤其是进入五月以来开启加速，Coreweave与英伟 达保持密切关系，英伟达持股占比3.86%，它是首个向公众提供基于NVIDIA GB200 NVL72实例的云服 务提供商，并且是首批部署 NVIDIA H100、H200和 GH200 高性能基础设施的云服务提供商之一。目前 CoreWeave共有32个数据中心，拥有超过25万个NVIDIA GPU，并得到超过260MW的电力支持。英伟 ...

Core Insights - Sify Technologies Limited has launched a Pay-per-use model to cater to the increasing demand for AI Cloud Services [1][3] - The company has expanded its portfolio of DGX-Ready Data Centers, now certified for up to 130 KW/rack capacity under NVIDIA's program [2] - The hourly pricing model includes hosting, power, and infrastructure costs, facilitating quicker deployment for GPU Cloud partners [3] Company Developments - Sify's new colocation pricing will be available at its certified data centers in Chennai, Noida, and Navi Mumbai [2] - The CEO of Sify Infinit Spaces Limited emphasized the company's extensive infrastructure and low-latency network connectivity to hyperscale clouds [4] - Sify aims to support the growing AI market in India by removing traditional barriers to AI adoption through its innovative pricing model [4] Industry Context - India is emerging as a key player in the global AI landscape, supported by a deep talent pool and advancing digital infrastructure [4] - Sify's pay-per-use model is positioned to enable global enterprises to leverage India's AI capabilities through scalable infrastructure [4] - The company has a significant presence, with over 10,000 businesses utilizing its services across more than 1,700 cities in India [7]