March 23, 2026 01:27 AM GMT Greater China Semiconductors | Asia Pacific AI Supply Chain Tracker: Key Investor Feedback from GTC/ OFC We visited NVIDIA's GTC and the OFC last week along with investors. GTC raised more questions than it answered, while investors generally left OFC more confident on optics. We remain positive on cloud semis and CPO outlook. We revise Aspeed's PT, given rack disaggregation suggests more BMC demand. GTC – New server rack designs and chip schedule ( Exhibit 1 ): Investors asked i ...

如果您希望可以时常见面，欢迎标星收藏哦~ 存储芯片供应商美光科技公司最新季度营收几乎翻了三倍，远超分析师预期，并且对当前季度也给 出了远超预期的业绩指引——然而，不知何故，其股价在盘后交易中仍然下跌了超过 4%。 该公司公布的未扣除薪酬等特定成本前的每股收益为12.20美元，远超华尔街预期的9.31美元。当 期营收增长194%，达到238.6亿美元，轻松超过市场普遍预期的207亿美元。 美光公司的盈利状况进一步提升。凭借这笔额外的收入，该公司本季度净利润达到137.8亿美元， 而去年同期仅为15.8亿美元。 该 公 司 业 绩 的 提 升 得 益 于 其 内 存 芯 片 需 求 的 飙 升 。 这 些 芯 片 已 成 为 英 伟 达 公 司 图 形 处 理 器 （GPU）的核心组件，而GPU正是人工智能工作负载的驱动力。当然，内存也广泛应用于几乎所 有其他类型的计算设备，包括个人电脑、智能手机、平板电脑和自动驾驶汽车。 美光是全球仅有的三家内存芯片制造商之一，也是唯一一家总部位于美国的。另外两家分别是三星 电子和SK海力士，均位于韩国。随着英伟达每一代GPU都需要更大的内存容量，全球内存供应变 得异常紧张，推 ...

Core Insights - The Nvidia GTC 2026 conference marks a significant reshaping of the supply chain dynamics, particularly with the introduction of the next-generation AI platform, Vera Rubin, and the roles of major chip manufacturers like Samsung, Micron, and Intel [1] Group 1: Nvidia's Product Developments - Nvidia's CEO Jensen Huang confirmed that the Groq 3 LPU will be manufactured by Samsung, continuing a prior agreement before Nvidia's acquisition of Groq for $20 billion [3] - The Groq 3 LPU is designed for high-speed inference and will be integrated into the Vera Rubin platform, expected to ship in the second half of 2026 [3] - The Vera Rubin platform will utilize TSMC's 3nm technology, featuring 72 GPUs and 36 CPUs, promising up to 10 times higher inference per watt at one-tenth the cost per token [2] Group 2: Memory and Supply Chain Dynamics - Micron announced that its 36GB 12-layer HBM4 has entered mass production in Q1 2026, with a pin speed exceeding 11 Gb/s and a bandwidth of over 2.8 TB/s, representing a 2.3 times improvement over HBM3E [4] - Micron's production ramp-up is expected to reduce supplier concentration in the HBM market, increasing pressure on existing suppliers like SK Hynix and Samsung [4] - Samsung's role has expanded from being an HBM supplier to also include logic chip manufacturing for Nvidia, enhancing its strategic position within the supply chain [3] Group 3: Intel's Strategic Collaborations - Intel confirmed that its Xeon 6 processors will support Nvidia's DGX Rubin NVL8 systems, offering a 2.3 times increase in memory bandwidth compared to the previous generation [5] - Intel is expected to collaborate with Nvidia on wafer-level packaging for the upcoming Feynman GPU, which will utilize TSMC's 1.6nm process technology [5] - The Feynman platform will introduce 3D chip stacking technology and custom HBM, differentiating Nvidia's AI data center offerings [5]

据官网信息，黄仁勋将于太平洋时间3月16日11:00-13:00(北京时间3月17日凌晨02:00-04:00)发表主题演 讲。届时黄仁勋将在圣何塞SAP中心登台，向开发者、分析师和媒体介绍英伟达未来一年的规划。 这位常以皮夹克穿着亮相的CEO，历来会在其主题演讲上密集发布新品与技术更新，本次活动预计也将 如此。 素有"AI界春晚"之称的英伟达年度开发者大会(GTC 2026)，将于周一加利福尼亚州圣何塞拉开帷幕，届 时英伟达首席执行官黄仁勋将发表主题演讲。本次大会将从3月16日持续至19日。 过去数月，英伟达展开了一系列收购与合作，覆盖芯片公司、软件厂商等。几乎可以肯定的是，本次大 会将披露这家AI芯片巨头如何将这些技术与能力整合进自身产品体系。 过去一年英伟达最受关注的交易之一是去年12月与AI芯片初创公司Groq达成的非独家协议，获得使用 后者推理技术的授权。根据协议条款，Groq创始人Jonathan Ross、总裁Sunny Madra及其他核心团队成 员将加入英伟达，共同推进授权技术的升级与规模化应用。 Groq研发的芯片被称为语言处理单元(LPU)，专为AI推理任务设计。该公司声称，LPU运行大 ...

Core Viewpoint - The upcoming NVIDIA GTC 2026 conference is expected to showcase significant product announcements and technological updates, particularly in AI and chip development, as CEO Jensen Huang will present the company's future plans [2][3]. Group 1: Product Announcements - NVIDIA is anticipated to reveal new products and technology updates during the GTC, including the integration of recent acquisitions and partnerships into its product ecosystem [3]. - A notable agreement with AI chip startup Groq allows NVIDIA to utilize Groq's inference technology, which claims to run AI models with ten times the efficiency of GPUs [3]. - NVIDIA may also introduce two new PC processors, N1 and N1X, designed for Windows laptops, which will be based on Arm architecture and target gaming scenarios [4][5]. Group 2: Software Developments - The company is expected to launch an AI agent platform named NemoClaw, enabling businesses to deploy intelligent agents within their systems [6]. - NVIDIA's focus on software capabilities is evident, with plans to release various open-source AI models applicable in multiple fields, including autonomous driving [6]. - The GTC will likely feature announcements related to AI and robotics, which could represent a multi-trillion-dollar market opportunity [6]. Group 3: Future Projections - Huang is expected to provide details on the upcoming Vera Rubin AI platform and the Vera Ultra set to launch in the second half of 2027, along with insights into the Feynman GPU expected in 2028 [5]. - The conference is anticipated to enhance market confidence in the sustained growth of the AI industry and the realization of incremental logic [6].

Core Viewpoint - The upcoming NVIDIA GTC conference is expected to signal a significant shift in the AI industry, particularly focusing on the transition from training to inference and adjustments in supply chain strategies [3][4][5]. Group 1: Key Signals from GTC - NVIDIA may leverage the integration of Groq technology to make a substantial entry into the AI inference market [5][6]. - The chip manufacturing process may shift from TSMC to Samsung, marking a potential break from TSMC's long-standing monopoly [5][7]. - The ecosystem for physical AI and open-source models is anticipated to expand further [5][10]. Group 2: Inference Market Focus - The AI industry is transitioning from a "training-first" approach to a "inference-driven" model, with NVIDIA's strategy being closely monitored [6]. - NVIDIA is expected to announce a new chip system that integrates Groq technology, which was acquired for approximately $20 billion [6]. - Groq's chips, known as Language Processing Units (LPU), are optimized for inference workloads, representing NVIDIA's first integration of another company's AI processor into its server architecture [6]. Group 3: Supply Chain and Client Developments - The Groq LPU is projected to be manufactured by Samsung in the latter half of the year, which could signify a shift in NVIDIA's reliance on a single supplier [7][8]. - OpenAI is expected to be one of the first customers for the new chip system, potentially utilizing it for AI tasks such as coding [8]. Group 4: Architectural Changes and Future Technology - The new system architecture will differ significantly from existing setups, featuring 256 Groq chips per rack, with Intel processors managing communication [9]. - NVIDIA is exploring deeper integration of LPU into future product roadmaps, including a potential single-chip solution combining Groq processors with next-generation Feynman GPUs [9]. Group 5: AI Application Ecosystem Expansion - NVIDIA's advancements in robotics and physical AI are gaining attention, especially in the context of the rapidly developing humanoid robot industry in China [10]. - The company is also progressing in the open-source model space, having released a 120 billion parameter model and planning to launch a new model with four times the parameters, which could lower AI inference costs and improve ROI [10]. Group 6: Long-term Industry Impact - The signals released at this GTC conference are likely to significantly influence the AI industry landscape by 2026 [11].

Core Insights - The upcoming NVIDIA GTC conference is expected to signal a strategic shift from training to inference in the AI industry, with significant implications for investors [1] - Key developments include the integration of Groq technology, a shift in supply chain dynamics, and the expansion of physical AI and open-source model ecosystems [1] Group 1: Shift to Inference Market - NVIDIA is transitioning from a "training-first" approach to a "inference-driven" strategy, responding to competition from companies like Cerebras that offer faster and cheaper solutions [2] - The company is expected to announce a new chip system that integrates NVIDIA and Groq technologies, following a $20 billion investment in Groq technology licenses [2] - Groq's chips, known as Language Processing Units (LPU), are optimized for inference workloads, marking NVIDIA's first integration of another company's AI processor into its server architecture [2] Group 2: Supply Chain Restructuring - The Groq LPU is anticipated to be manufactured by Samsung in the second half of the year, representing a significant shift away from NVIDIA's long-standing reliance on TSMC for chip production [3] - This change may be temporary, as future LPU production could return to TSMC to ensure tighter integration with NVIDIA's upcoming AI chips [3] - OpenAI is expected to be one of the first customers for the new chip system, which may be utilized for AI-related tasks such as coding execution [3] Group 3: Architectural Changes and Future Technology Roadmap - The new system architecture will feature 256 Groq chips per rack, with Intel processors managing communication, indicating that the integration of LPU with existing systems is still in progress [4] - NVIDIA is exploring deeper integration of LPU into its future product roadmap, potentially merging Groq processors with the next-generation Feynman GPU to enhance performance and reduce costs [4] Group 4: Expansion of Physical AI and Open-Source Models - NVIDIA's focus on the AI application ecosystem is highlighted by its advancements in robotics and physical AI, particularly in the context of the rapidly growing humanoid robot industry in China [6] - The company has released a 120 billion parameter model, Nemotron 3 Super, and plans to introduce a new model, Nemotron 4 Ultra, with four times the parameters, which could lower AI inference costs and improve ROI for enterprises [6] - The signals from this GTC conference are likely to significantly influence the AI industry landscape by 2026 [6]

Core Viewpoint - Nvidia is positioned favorably ahead of the GTC 2026 conference, with analysts expressing optimism about its stock performance and potential announcements related to AI technologies and partnerships [1]. Group 1: Stock Performance and Analyst Ratings - Bank of America Securities has maintained a "Buy" rating for Nvidia with a price target of $300, highlighting key areas for investors to monitor [1]. - Nvidia's stock has seen modest gains of 2.5% over the last six months, underperforming compared to the iShares Semiconductor ETF, which has returned approximately 30% in the same period [3]. - The company's PEG ratio has remained below 1x for the last two quarters, indicating that the stock is undervalued relative to its future earnings potential [4]. Group 2: Financial Health and Valuation - Nvidia has a significant cash reserve of $62.56 billion and a low debt-to-equity ratio of 7.25%, suggesting strong financial health and capacity for continued investment in R&D [5]. - The company's price-to-book ratio has decreased from 51.65x to 28.74x, reflecting a more attractive valuation as it has grown in size [4]. - Over the past 12 months, Nvidia has generated a levered free cash flow of $58.13 billion, with a return on equity exceeding 100% [5]. Group 3: Future Developments and Focus Areas - Key areas of focus for Nvidia include updates on the Feynman GPU product pipeline expected in 2028, advancements in low-latency and inference products, and developments in networking and quantum computing [7]. - Analysts are particularly interested in insights regarding the projected Rubin ramp-up and Nvidia's 102.4T Spectrum-6 switch pairs, as well as a potential joint announcement with Intel on a custom CPU [1].

Core Insights - Nvidia is set to host its GPU Technology Conference, where it is expected to unveil new products and provide updates on its roadmap, including the Rubin Ultra line of chips and the Feynman GPU [1][5] - The event is seen as a crucial opportunity for Nvidia to regain investor confidence after a challenging start to the year, despite strong past performance [2] - Analysts are cautious, noting that Nvidia faces high expectations and a challenging sentiment in the AI sector, which may impact its ability to impress investors [2][3] Product Developments - Anticipated announcements include the Rubin Ultra chips and a new AI agent platform called "NemoClaw" [1] - The conference may also highlight advancements in Nvidia's software related to robotics and AI applications in the physical world [1] Market Sentiment - Despite a weak stock performance, Wall Street analysts remain largely positive, with 12 out of 13 analysts rating Nvidia as a "buy," suggesting a potential upside of 45% from its recent closing price [4] - Nvidia's stock has seen a decline of about 3% in 2026, although it has regained approximately half of its value compared to the same time last year [4]

Key Takeaways Nvidia is set to kick off its weeklong GPU Technology Conference on Monday, where the AI chipmaker is expected to offer more details on its latest products. CEO Jensen Huang is set to give a keynote address on Monday at 2 p.m. ET. The week ahead could be a big one for AI chip leader Nvidia—and the AI trade. Nvidia (NVDA), the world's most valuable company, is set to kick off its weeklong GPU Technology Conference Monday. The company is widely expected to unveil new products and give m ...