【TechWeb】3月19日消息，据外媒报道，在近几年的生成式人工智能热潮中，黄仁勋担任CEO的英伟 达是一大受益者，他们的算力芯片被各大厂商大量购买，他们的业绩也大幅提升，营收已连续11个财季 创下新高，在截至今年1月25日的2026财年第四财季已经达到了681.27亿美元。 而从市场研究机构的报告来看，在生成式人工智能热潮中获益颇丰的英伟达，已连续两年是全球营收最 高的半导体供应商，去年的营收遥遥领先。 【TechWeb】3月19日消息，据外媒报道，在近几年的生 成式人工智能热潮中，黄仁勋担任CEO的英伟达是一大 受益者，他们的算力芯片被各大厂商大量购买，他们的 业绩也大幅提升，营收已连续11个财季创下新高，在截 至今年1月25日的2026财年第四财季已经达到了681.27亿 美元。 而从市场研究机构的报告来看，在生成式人工智能热潮 中获益颇丰的英伟达，已连续两年是全球营收最高的半 导体供应商，去年的营收遥遥领先。 市场研究机构的报告显示，在2024年和2025年，英伟达 的营收分别为973.95亿美元、1503.01亿美元，均是营收 最高的半导体供应商。 在2024年和2025年，营收第二高的半导体供应 ...

GTC 2026现场，老黄直接怒怼玩家：他们完全不懂AI！ 啥情况？ 原因是周一英伟达刚发布新一代图形技术DLSS 5，本该是一次"性能革命"，结果却遭到了游戏圈集体抵制。 它能让《生化危机：安魂曲》中的金发女郎格蕾丝·阿什克罗夫特变成这样： 也能让里昂·肯尼迪变成这样： 看起来颜值爆表是吧？但玩家们丝毫不买账。 他们表示：我买的是艺术家制作的游戏，现在怎么全是AI生成的网红脸？ 但老黄的说法也同样强硬：DLSS 5是图形领域的"GPT时刻"。 这不是后期处理，不是帧级别的后期处理，而是几何级别的生成控制。 面对玩家的强烈反对，老黄却变本加厉地强调：DLSS 5"不会改变艺术控制"，开发者仍然可以"微调生成式AI"以匹配他们的风格。 DLSS 5：给游戏加上AI滤镜 要理解这场争议，先得搞清楚DLSS 5和之前有什么不同。 DLSS（深度学习超级采样），是英伟达开发的一套AI驱动的图形渲染技术。 从1.0到4.0，其核心逻辑一直是"超分辨率"和"帧生成"——用AI把低分辨率画面放大，或者在两帧之间插入AI生成的过渡帧，目的是提升性能、降低显卡 负载。 简单来说，就是让AI帮你"脑补"出更清晰、更流畅的游戏画面 ...

Sometimes You Don’t Want A GPU: Groq Cofounder Explains Whirlwind Deal With NvidiaThe Christmas Eve agreement—billed as Nvidia’s biggest deal in its three-decade history—landed at a precarious moment for Groq. Now Nvidia is betting on Groq’s inference-speed tech inside a newly announced chip platform. https://t.co/JOA2PtXZzF ...

Nvidia reopens China channel, but H200 flows remain policy-constrained Nvidia reopens China channel, but H200 flows remain policy-constrained Artificial intelligence Author Devesh Kumar Mar 19, 2026, 10:27 AM Nvidia is restarting production of its H200 artificial intelligence processors for Chinese customers after months of uncertainty, with CEO Jensen Huang confirming that purchase orders are in place and manufacturing is ramping back up. The move points to a partial reopening of a key market under tighter ...

来源 | 央视新闻 在3月19日商务部举行的例行发布会上，有记者提问，"相关报道称中国政府已经批准了部分公司采购英 伟达的H200芯片，请问是否有更多细节可以透露以及有何评论？" 商务部新闻发言人何咏前回应 称："我不了解你提到的情况。" 编辑 | 杨家瑞 ...

Core Viewpoint - Bank of America Securities analyst Vivek Arya maintains a bullish outlook on Nvidia, emphasizing its expanding role in the AI ecosystem and setting a price target of $300 [1] Group 1: Data Center Opportunity - Nvidia's projected data center revenue is expected to exceed $1 trillion for the period of 2025–2027, which does not account for contributions from CPUs, storage systems, and newer rack offerings, indicating a potential 50% increase in total opportunity as AI adoption grows [2] Group 2: Profit Potential in AI Workloads - The cost of generating AI outputs has been reduced to approximately $6 per 1 million tokens, with Nvidia entering ultra-low-latency workloads through its LPX systems, which could capture about 25% of the market and yield higher profitability [3] Group 3: Long-Term Growth Roadmap - Nvidia's roadmap includes advancements in combining copper and optical technologies and a shift towards fully advanced optical solutions, with an increasing share of AI workloads expected from enterprise and sovereign customers, supporting long-term growth [4] Group 4: Technical Analysis - Nvidia is currently trading 2.3% below its 20-day simple moving average of $184.44 and 3.1% below its 100-day SMA of $185.89, with a 12-month stock gain of 53.51%, indicating a short-term consolidation phase [5] - The RSI is at 45.31, suggesting neutral momentum, while the MACD at -1.1357 indicates bearish pressure affecting the near-term trend [6] Group 5: Earnings & Analyst Outlook - The next significant catalyst for Nvidia's stock is the estimated earnings report on May 27, 2026 [7] Group 6: Analyst Consensus & Recent Actions - Nvidia carries a Buy rating with an average price target of $280.96, with recent analyst upgrades raising targets to $287.00 and $325.00 [8] - Key resistance is noted at $185.00 and key support at $178.00, with an EPS estimate of $1.73 and revenue estimate of $78.69 billion, reflecting significant year-over-year growth [8] Group 7: ETF Exposure - Nvidia's significant weight in ETFs means that any major inflows or outflows could lead to automatic buying or selling of the stock [9] Group 8: Price Action - During premarket trading, Nvidia shares were down 0.50% at $179.50 [10]

Core Viewpoint - Nvidia is expanding its business beyond GPUs, positioning itself as a comprehensive provider of AI infrastructure, including data center accelerators, hardware, and software solutions for physical AI applications like robotics and autonomous driving [3][4]. Group 1: Product Expansion - At the GTC conference, Nvidia introduced a variety of products, including data center accelerators, networking products, and open-source models, indicating a shift towards a broader AI infrastructure role [3]. - The introduction of the Groq 3 and Groq 3 LPX chips, which enhance the performance of Nvidia's Rubin platform, signifies a diversification in Nvidia's product offerings beyond traditional GPUs [7]. - The Groq 3 LPX chip can increase inference throughput by 35 times per megawatt when used with Rubin CPU and GPU, showcasing significant performance improvements [7]. Group 2: Chip Heterogeneity - Nvidia's strategy includes integrating LPU (Language Processing Unit) technology with GPUs to address the growing demand for faster inference speeds in large models, indicating a trend towards chip heterogeneity in AI workloads [10][11]. - Ian Buck emphasized that while GPUs will continue to dominate current AI applications, the combination of LPU and GPU will be crucial for next-generation AI workloads, particularly those involving trillion-parameter models [10]. - The industry is moving towards a heterogeneous computing environment, where different types of chips are needed for various workloads, as highlighted by AMD's collaboration with Meta to design semi-custom chips [10][11]. Group 3: Physical AI Development - Nvidia is making significant strides in physical AI, launching the Isaac simulation framework and the Cosmos model for robotics, which aims to unify synthetic world generation and physical AI reasoning [15][18]. - The company is focusing on open-source technologies to foster collaboration in the development of physical AI, as it believes that no single company can achieve this alone [15][19]. - Rev Lebaredian noted that the challenges in robotics are multifaceted, requiring advancements in hardware and software to make robots more functional and accessible [19].

行业点评报告 · 通信行业 英伟达引领 AI 基建时代,LPU 拓展推理边界 2026年03月18日 核心观点 通信行业 推荐 维持评级 分析师 赵良毕 ☎: 010-8092-7619 区: zhaoliangbi_yj@chinastock.com.cn 分析师登记编码：S0130522030003 开标 ☎: 0755-8347-9312 网: hongye_yj@chinastock.com.cn 分析师登记编码：S0130523060002 相对沪深 300 表现图 2026年03月18日 100% 50% 资料来源：Wind，中国银河证券研究院 相关研究 1.【银河海外算力】行业周报_OFC 即将举办，重点 聚焦硅光子学等板块 2. 【银河通信】行业周报_光纤价格增长显著, OpenClaw 再增算力需求 3. 【银河通信】行业点评报告: 新基建超前布局, 十 五五新质生产力蓬勃发展 www.chinastock.com.cn 证券研究报告 请务必阅读正文最后的中国银河证券股份有限公司免责声明 0 事件: 2026年 3 月 17 日凌晨,英伟达 2026年 GTC 大会拉开帷幕,创始人 兼 CE ...

Core Viewpoint - Nvidia's stock remains stable following its GTC conference, yet Wall Street analysts continue to express optimism, with notable support from Elon Musk [1] Group 1 - Nvidia's stock performance is not significantly affected by the recent GTC conference [1] - Analysts on Wall Street are praising Nvidia's developments and potential [1] - Elon Musk has publicly supported Nvidia, adding to the company's positive sentiment [1]

Investment Rating - The report maintains a "Recommended" rating for the electric power equipment industry [1]. Core Insights - The AI industry is transitioning from generative AI to inferential AI, with a significant increase in computational demand, approximately 1 million times over the past two years [2]. - Key performance indicators for AI factories include throughput efficiency and inference speed, with token efficiency being crucial for power efficiency [2]. - NVIDIA anticipates that the revenue from its Blackwell and Rubin flagship chip product lines will exceed $1 trillion by 2027, a significant increase from the previously projected $500 billion [2]. Summary by Sections Industry Progress - The AI industry is evolving towards inferential and intelligent AI, with inference and channel training becoming core computational demands [2]. - NVIDIA has defined 2025 as the "Year of Inference," focusing on optimizing AI inference processes and reducing infrastructure costs [2]. Product Developments - The Vera Rubin full-stack AI computing platform was officially launched, featuring several new chips and configurations [2]. - The computational power of the Vera Rubin platform is expected to increase by 40 million times over the next decade [2]. - Deployment efficiency for Vera Rubin has improved significantly, reducing installation time from two days to two hours [3]. Power Supply Developments - The cabinet power supply for the Vera Rubin NVL72 is expected to utilize four Powershelf units, each with a power capacity of 110KW, totaling 440KW, which is over 60% higher than previous models [2][3]. - NVIDIA plans to implement 800V high-voltage DC power supply systems to reduce data center PUE below 1.1 [2]. Liquid Cooling Developments - The Vera Rubin platform employs a 100% liquid cooling design, which is expected to enhance energy efficiency and reduce cooling costs [3]. - Future releases of higher power chips are anticipated to drive technological upgrades in liquid cooling solutions [3]. Investment Recommendations - Suggested companies to watch include those involved in cabinet power supplies, external power supplies, liquid cooling solutions, and data center storage [3].