什么是 英伟达？

NVIDIA unveils RTX Spark AI PC chip with Microsoft, redefining Windows PCs as native agent platforms supporting local LLMs, gaming, and pro workflows — marking a new era of personal computing.

Crossover Intelligence topped the World Arena Track 2 with its DSCFuncWorld, outpacing the second-place model by a significant margin and validating end-to-end data generation, strategy training, and task execution capabilities.

Future AI compute power will be dominated by inference at 70%, with communication energy consumption being the main bottleneck in data centers, and physical AI development depends on high-quality real-world data.

大模型参数规模突破T级得益于新型架构如Gated Delta Networks和Mamba的创新，英伟达、Kimi、Qwen等均采用混合架构优化性能。

英伟达季度营收逼近1000亿美元，Rubin Ultra架构未延期，Anthropic等客户转向GPU算力，AI算力市场格局生变。

英伟达推出ENPIRE框架，让AI agent自主完成机器人研究，成功率达99%。

英伟达推出开源框架 Polar，显著提升 Codex 等智能体的性能和效率。

Future AI inference will consume 70% of computing power, communication energy consumption can reach hundreds of times that of computation, and high-quality data and edge computing will become key bottlenecks.

NVIDIA GEAR Lab introduces DreamDojo and DreamZero world models, advancing embodied intelligence.

The article discusses new trends in the AI field, emphasizing the importance of recursive AI architecture and highlights Tian Yuandong's collaboration with top teams to drive AI evolution.

Wailiang Technology's BabyAlpha A3 robot breaks NVIDIA's computing monopoly, achieving a qualitative leap in perception and computing capabilities for consumer robots.

Under Huang Renxun's leadership, NVIDIA invested $17.5 billion in private firms and infrastructure funds in the past fiscal year, and another $40 billion in early 2026—averaging over $298 million per day, with $30 billion going to OpenAI alone.

英伟达Jim Fan宣布VLA路线过时，提出新范式WAM，代表作为DreamZero，预计2040年实现机器人自主设计制造，置信度95%。

Jensen Huang said at the graduation ceremony that AI won't replace people, but those who use it will surpass others.

Four tech giants announced capital expenditures of $700 billion for 2026, all directed at AI infrastructure, but market reactions are polarized. Behind the investment boom is a game between supply-constrained markets and fulfilling real demand.

SpaceX launches $1.25 trillion IPO, Meta uses employee mouse/keyboard data for AI training, NVIDIA's Q1 data center revenue surges 92% to $75.2B, GoPro explores sale or merger.

OpenAI announces robotics expansion with hiring drive; Anthropic files confidential IPO, possibly ahead of OpenAI; major automakers report strong May sales; Tianya community reboots but crashes under traffic; Xianyu apologizes for AI mislisting relics; Nvidia's Vera Rubin platform enters mass production for agent AI factories; Doubao to launch paid tier in late June.

The article mainly discusses Guanglun Intelligent's collaboration with Google and NVIDIA to define physical AI simulation standards, but lacks specific technical details and practical content.

Jensen Huang urges students to embrace AI, calling it a once-in-a-lifetime opportunity that will reshape education and employment.

NVIDIA invests $2.1 billion in IREN to build 5GW AI data centers, signaling strategic focus on global AI infrastructure.

Apple WWDC theme exposed, NVIDIA releases personal AI PC with stock up 6%, Yushu IPO approved setting 73-day lightning record.

NVIDIA's open-sourced Skill security scanner lacks technical details, low information density, not worth deep reading for engineers.

AI voice company ElevenLabs increased its ARR from $350M to $500M in six months and secured investment from NVIDIA.

This is a morning news digest from GeekPark covering three tech stories: Musk's lawsuit against Altman resulting in Musk's loss, Kujiale's partnership with Insta360 to democratize 3D content creation, and NVIDIA's Q1 revenue potentially exceeding expectations. However, the full article content could not be accessed due to WeChat platform verification restrictions.

文章内容为英伟达 CEO 黄仁勋关于品格与智慧的引言，缺乏技术深度和实用信息。

文章标题提及红杉、英伟达投资前DeepMind研究员创立的Ineffable，种子轮估值51亿美元，但正文内容无法正常访问，信息无法验证。