# (77) NVIDIA GTC Taipei 2026 Keynote | Full Replay - YouTube
- Transcript ID: `wsp6ainirsy`
- Source tier: `youtube_auto`
- Original language: `en_or_source_language_unlabeled`
- Translation status: `agent_translated_reviewed` (227/227 chunks)
- Original source: https://www.youtube.com/watch?v=wSp6AiNIrsY
- Raw evidence: `data/transcripts/youtube/wSp6AiNIrsY.json`
> 中文是派生翻译，英文是留存的源文本。中文未复核时，以英文和原始音视频为准。
## 00:00:02
**中文**
[音乐] >> 智能就是这样产生的：在一种新型工厂中。>> [音乐] >> 它生成 token，也就是 AI 的构建模块。token 开启了一个新前沿，把数据转化为知识、推理与行动；它们揭示出我们过去无法看见的复杂模式。
**English / Original**
[music] >> This is how intelligence is made. A new kind of factory. >> [music] >> Generator of tokens. The building blocks of AI. Tokens have opened a new frontier. [music] Turning data into knowledge, reason, action. They reveal patterns in complexity. >> [music] >> We could never see.
## 00:00:47
**中文**
>> [音乐] >> 它们映射我们的城市，守护我们的安全；>> [音乐] >> 托举我们飞越城市上空。>> [音乐] >> token 帮助机器人向我们学习，>> [音乐] >> 与我们并肩工作，去往人类无法抵达之处。[音乐]
**English / Original**
>> [music] >> Mirror our cities. To keep us safe. >> [music] >> And lift us high. Above them. >> [music] >> Tokens help robots learn from us. >> [music] >> Work [music] alongside us. They go where we cannot. [music]
## 00:01:30
**中文**
向我们伸出援助之手。 [音乐]并缩小希望和治愈之间的差距。让我们呼吸更轻松。最小的心脏跳动得更强劲。 >> [音乐] [音乐] [音乐] >> token 正在帮助我们开辟新天地
**English / Original**
Lending us helping hands. [music] And closing the gap between hope and healing. So that we breathe easier. And the smallest hearts beat stronger. >> [music] [music] [music] >> Tokens are helping us break new ground
## 00:02:10
**中文**
>> [音乐] >> 规模从未尝试过。 >> [音乐] [音乐] >> 这样我们就可以到达>> 星云1号分离已确认。 >> 无限 >> [音乐] >> 及以上。 >> [音乐] >> 我们一起迈出下一个[音乐]伟大飞跃，进入光明的新未来>> [音乐] >> 为全人类而建。
**English / Original**
>> [music] >> on a scale never attempted. >> [music] [music] >> So we can reach >> Star Cloud 1 separation confirmed. >> to infinity >> [music] >> and beyond. >> [music] >> Together, we take the next [music] great leap into a bright new future >> [music] >> built for all mankind.
## 00:02:58
**中文**
>> [音乐] >> 台北是一切的开始。 >> [音乐] >> 欢迎上台，NVIDIA 创始人兼首席执行官黄仁勋。 >> [音乐] >> 欢迎来到GTC [音乐]台湾。很高兴见到你们大家。回家真好。我带了我的
**English / Original**
>> [music] >> And here in Taipei is where it all begins. >> [music] >> Welcome to the stage, NVIDIA founder and CEO Jensen Huang. >> [music] >> Welcome to GTC [music] Taiwan. So great to see all of you. Very good to be home. I brought my
## 00:03:40
**中文**
父母回家。我的父母在哪里？大家把掌声送给我的爸爸妈妈。 >> [掌声] [掌声] >> 女士们、先生们，为我们的赛前表演超级明星们鼓掌。 >> [掌声] >> 看看他们多可爱。台湾的巨星。呃，今天这里有很多人。我们现在正在向台湾各地的其他 70 个观看团体广播此内容。
**English / Original**
parents home. Where are my parents? Everybody give round of applause to my mom and dad. >> [applause] [applause] >> And a round of applause for our pre-game show superstars, ladies and gentlemen. >> [applause] >> Look how adorable they are. The superstars of Taiwan. Uh there are so many of you here today. We are broadcasting this right now to 70 other watch parties across Taiwan.
## 00:04:21
**中文**
70 个[轻笑]不同的会议同时举行。大家都在关注这个主题演讲。我们有很多话要告诉你，我有很多合作伙伴要感谢。台湾的生态系统已经变得如此庞大，令人难以置信。大多数时候，当人们想到生态系统时，他们会想到我们的软件堆栈。他们考虑的是 NVIDIA 构建的计算系统之上的开发者生态系统。
**English / Original**
70 [snorts] different conferences are going at the same time. Everybody is watching this keynote. We have so much to tell you, and I have so many partners to thank. It is incredible how large our ecosystem in Taiwan has become. Most of the time, when people think about ecosystem, they think about our software stack. They think about the developer ecosystem above the computing systems that NVIDIA builds.
## 00:04:48
**中文**
但 NVIDIA 的生态系统从上游一直延伸到我们在台湾的所有供应链（这一切都是从这里开始），一直延伸到下游数据中心并最终到达最终用户。今天，我们将讨论几乎所有的生态系统。有很多人要感谢。我喜欢这里的生态系统。我的意思是，这里有很多公司，还有一些我最喜欢的生态系统合作伙伴>> [掌声]
**English / Original**
But NVIDIA's ecosystem spans all the way upstream to all of our supply chain here in Taiwan, where it all begins, and downstream all the way to data centers and eventually to end users. Today, we're going to talk about almost all of the ecosystem. There's so many people to thank. I love my ecosystem here. I mean, there's so many companies here, and some of my favorite ecosystem partners >> [applause]
## 00:05:51
**中文**
>> 台湾如此丰富的生态系统。最丰富的生态系统，全球最好的供应链生态系统。难以置信。好的，谢谢大家来到这里。今年，今年我们的业务共同增长得令人难以置信。事实上，昨晚有人告诉我，台湾的GDP每年将增长近10%。 >> [掌声] >> 难以置信。嗯，我们有很多话要谈。我们开始吧。
**English / Original**
>> So many Taiwan's rich ecosystem. The richest ecosystem, the world's best supply chain ecosystem. Unbelievable. Well, thank you all for being here. And uh this year this year our businesses together are growing incredibly. In fact, somebody told me last night that the annual GDP of Taiwan is going to grow almost 10%. >> [applause] >> Unbelievable. Well, we have a lot to talk about. Let's get going.
## 00:06:30
**中文**
两年前，当我在这里时，我开始与大家谈论人工智能如何从生成式人工智能以及即将到来的其他人工智能浪潮中转变。人工智能的下一波浪潮是智能体 AI。今天我们可以说智能体 AI 已经到来，有用的人工智能已经到来。现在，这意味着什么？这是 GitHub。这当然是智能体 AI 的第一个应用是软件编码。
**English / Original**
Two years ago when I was here, I started to talk to you about how AI has moved from generative AI and the other waves of AIs that are coming. The next wave of AI was agentic AI. And today we can say that agentic AI has arrived, that useful AI has arrived. Now, what does this mean? This is GitHub. This is of course one of the first applications of agentic AI is software coding.
## 00:06:58
**中文**
这是最有价值的职业之一，也是一个规模惊人的生态系统：全球有三四千万名专业软件开发者，此外可能还有数亿学生、爱好者等。但我们先看靠写代码谋生的三四千万名软件开发者，GitHub 基本代表了他们。pull request 是他们下载软件、修改代码后提出合并请求；commit 则是把修改提交回去。
**English / Original**
One of the most valuable professions. Incredibly large ecosystem. 30 million, 40 million professional software developers. Probably another couple of hundred who are students and enthusiasts and so on and so forth. But say 30, 40 million software developers in the world code for a living. And this represents most of them. This is GitHub. The pull request is when they download software, they modify it, and commit is when they push it back up.
## 00:07:32
**中文**
明白吗？看这张图：2023 年有 3 亿次 commit，2024 年 4 亿次，2025 年 5 亿次；而在 2026 年最初几个月里，这个数字已接近原来的三倍。这意味着什么？三千万名软件开发者代表约 3 万亿美元的 GDP，正在产出三倍的——这就是他们
**English / Original**
Okay? And so if you could look at this in 20 23 the number of commits was 300 million. 2024 400 million. 2025 500 million commits. In the first few months in the first few months of 2026, it has nearly tripled. Now, what does that mean? 30 million software developers representing about 3 trillion dollars worth of GDP producing three That's what they're
## 00:08:12
**中文**
获得的报酬。每年 3 万亿美元的薪资，推动其他行业产生经济增长。假设全球约 100 万亿美元的产业都受到这部分劳动影响；刚才说 30 亿美元是口误，应该是 3 万亿美元。如今，这 3 万亿美元薪资对应的产出已接近过去的三倍。
**English / Original**
paid. 3 trillion dollars worth of salaries per year, which is generating economic growth for the rest of the industries. Say 100 trillion dollars of the world's industries is impacted is generated by 3 billion dollars worth of salary. That 3 trillion dollars Excuse me, 3 trillion. That 3 trillion dollars worth of salary is now producing nearly three times as much output.
## 00:08:42
**中文**
这实际上是 3 万亿美元的工资带来了 9 万亿美元的生产力。这有什么意义吗？差异绝对非同寻常。这就是潜力。这是人工智能的承诺。工程师、软件工程师的数量实际上正在增加。人们谈论人工智能减少工作岗位。完全是废话。
**English / Original**
It's effectively a 9 trillion dollar productivity from 3 trillion dollars of salaries. Does that make any sense? The difference is absolutely extraordinary. This is the potential. This is the promise of AI. The number of engineers, software engineers, is actually increasing. People talk about AI reducing jobs. Complete nonsense.
## 00:09:06
**中文**
它导致雇用更多的软件工程师，原因很简单。如果您可以雇用一名软件工程师并且可以产生价值 9 万亿美元的生产性工作，那么您为什么不想雇用更多的软件工程师呢？如果这条线是平坦的，那么显然人们会雇佣更少的软件工程师。但由于产出如此令人难以置信，人们希望雇用更多的软件工程师。这很快就会在我们的经济中显现出来。
**English / Original**
It's causing more software engineers to be hired and the reason for that is very simple. If you can hire a software engineer and you could generate 9 trillion dollars worth of productive work, why wouldn't you want to hire more software engineers? If that line was flat then obviously people will hire fewer software engineers. But because the output is so incredible, people want to hire more software engineers. This is going to show up in our economy somehow soon.
## 00:09:37
**中文**
所以，第一件事是有用的人工智能已经到来。那么，从行业的角度来看，这意味着什么呢？从行业的角度来看，这意味着token现在的需求量非常大。因为如果你能做到这一点，你就会想要生产更多。因为token现在是盈利单位。token现在是可盈利的收入单位。
**English / Original**
And so, the first thing is useful AI has arrived. Now, what does that mean from the industry's perspective? From the industry's perspective, that means that tokens are now in extraordinary demand. Because if you could do this, you're going to want to produce more of it. And because tokens are now profitable units. Tokens are now prof- profitable units of revenues.
## 00:10:02
**中文**
因为现在已经盈利了，人工智能公司想要建立更多的token，产生更多的token，建立更多的AI 工厂，这就是台湾计算需求飙升的原因。也正是因为如此，你们才这么忙，你们的生意做得这么好。事实上，这看起来像是你的一些股票价格。
**English / Original**
Because it is now profitable, the AI companies want to build a lot more tokens, generate a lot more tokens, build more AI factories, which is the reason why compute demand here in Taiwan has skyrocketed. It is precisely the reason why all of you are so busy and your businesses are doing so well. In fact, that looks like some of your stock price.
## 00:10:30
**中文**
>> [掌声] >> 计算模式已经改变。一切都变了。所以，第一个想法是有用的人工智能已经到来。人工智能现在是利润创造者。人工智能现在是GDP的创造者。而其背后是一种全新的计算模式。不仅仅是一个大型语言模型，而且是一个智能体。今天，我们要讨论的几乎所有内容都将基于此。那么，让我花一点时间向您展示我在说什么。这里面有一个这是一个智能体。
**English / Original**
>> [applause] >> The compute pattern has changed. Everything has changed. So, the first idea is that useful AI has arrived. AI is now a profit generator. AI is now a GDP generator. And behind it is a whole new kind of computing pattern. Not just a large language model, but an agent. Today, almost everything we're going to talk about is going to be based on this. So, let me take a quick moment and show you what I'm talking about. Inside in this is a This is an agent.
## 00:11:06
**中文**
这是一个智能体应用程序。在过去，这就是应用程序。这将是代码。这就是操作系统。应用程序，在操作系统内的应用程序内运行的代码。今天，它是智能体，由一个大型语言模型或许多位于编排框架内的语言模型组成，该编排框架帮助它协调它以完成富有成效的工作。
**English / Original**
It's an agent application. In the old days, this would be application. This would be code. And this would be operating system. Application, code running inside an application inside an operating system. Today, it is agent, which consists of a large language model or many sitting inside a harness and that harness helps it orchestrates it to do productive work.
## 00:11:39
**中文**
这是输入。输入到来后，智能体必须理解、观察、推理、行动并使用工具。工具可以是电子表格、网页浏览器、数据处理引擎或数据库引擎。编排框架负责调度每一次信息流转：处理上下文、理解正在发生什么、推理应该做什么、制定计划并执行。这个编排路径由
**English / Original**
This is the input. When that input comes, it has to understand, observe, reason, act, use tools, use tools. That tool could be a spreadsheet, web browser, a data processing engine, database engine, for example. This is orchestrated this harness orchestrate this routing of information every single time it touches either processing the context, understanding what is happening, reasoning about what to do, coming up with a plan that you can act that it acts on. That orchestration path is orchestrated by
## 00:12:22
**中文**
一些软件。
**English / Original**
some software.
## 00:12:23
**中文**
所以这本质上是一个智能体。它涉及短期记忆，称为工作记忆，长期记忆，就像我们一样，我们有长期记忆。因此内存管理系统非常重要。整个系统称为智能体。大语言模型用于进行思考，而编排框架将所有内容连接在一起，就像操作系统一样。
**English / Original**
And so this is fundamentally a agent. It deals with short-term memory, called working memory, long-term memory, just like we do, we have long-term memory. And so the memory management system is incredibly important. This entire system is called an agent. The large language model is used to do the thinking and the harness connects everything together, just like an operating system.
## 00:12:52
**中文**
明白吗？这就是新的计算模型，也就是智能体。它能够完成非凡的工作。重大突破来自两者的汇合：大语言模型如今已经能够出色地思考、推理、规划和使用工具；与此同时，我们有了能够管理记忆、执行编排并调用工具的框架。两者结合，让我们能够做出惊人的事情。下面举几个例子。
**English / Original**
Okay? And so this is the new computing model and this is what an agent. It could do incredible things. This is the big breakthrough. The simultaneous convert the convergence of large language models that are now able to do a really good job thinking, reasoning, planning, using tools, and the fact that we have now these harnesses that manages memory, the orchestration, uses tools, we can now do amazing things. Let me give you some example.
## 00:13:22
**中文**
这是这是一个提示。这是提示。这是生成的代码，然后就出来了。这是输入。这是输入，这是输出。各位小伙伴你们觉得怎么样呢？这真是太神奇了，对吧？ >> [掌声] >> 老兄，我们在这里使用Claude Code，但 Codex 也做得非常出色。这是另一个例子。
**English / Original**
This is This is a prompt. This is the prompt. This is the code that is generated and this comes out. This is the input. This is the input and that's the output. Do you guys What do you guys think? It's pretty amazing, right? >> [applause] >> Dude, we use cloud code here, but Codex it does an incredible job as well. Here's another example.
## 00:13:52
**中文**
这是输入，创建一个 GIF NVIDIA 黑底绿点散布形成台湾 101 建筑 morph 2 GTC type A 2 2026 morph 2 NVIDIA 眼睛标志，然后散布并重复。正确的？所以，你看到了。这就是提示。这是下一篇。我丢失了遥控器电池夹。看起来像这样。创建 CAD 文件。它使用一个工具。创建一个 CAD 文件，准备进行 3D 打印以创建新的文件。
**English / Original**
This is the input, create a GIF NVIDIA green dots on black scatter form Taiwan 101 building morph 2 GTC type A 2 2026 morph 2 NVIDIA eye logo then scatter and repeat. Right? So, you saw that. That was the prompt. Here's the next one. I lost my remote control battery clip. It looks like this. Create a CAD file. It uses a tool. Create a CAD file ready for 3D printing to create a new new one.
## 00:14:26
**中文**
有道理吗？这是现在新的计算模式。我们过去常常启动应用程序，单击并键入，但现在我们将其替换为向人工智能解释我们想要什么、我们的意图，然后人工智能生成代码或使用工具并产生必要的输出。这就是计算机未来的工作方式。这就是智能体 AI。
**English / Original**
Make sense? This is now the new computing pattern. Whereas we used to launch an application, click and type, we now replace that with explaining to the AI what we want, our intent, and the AI generates the code or uses tools and produce the necessary output. This is how computers are going to work in the future. This is agentic AI.
## 00:14:57
**中文**
两年来，我们一直朝这个方向建设，如今它已经到来。其中一个重大突破当然是工具使用。很多人曾说：“黄仁勋，AI 要来了，智能体 AI 要来了，所以所有软件公司都会倒闭。”我说恰恰相反：世界将拥有如此多的智能体，使用者数量将不再受人类人口规模限制。
**English / Original**
For 2 years we've been building towards this and now it has arrived. Now, one of the big breakthroughs, of course, is tool use. A lot of people have said, "You know, Jensen, AI's coming, agentic AI's coming, therefore all of the software companies are going to go out of business." I said it's exactly the opposite because there are going to be so many agents, the world is no longer at by the number of people.
## 00:15:22
**中文**
因此，这些智能体将使用比以往更多的工具。现在其实是创办软件公司的绝佳时代，但软件必须以智能体能够调用的方式呈现。这是一项重大突破。大家知道，NVIDIA 的宝藏就是全部 CUDA 库，我把它们称为 CUDA-X 库；这正是 NVIDIA 的宝藏。
**English / Original**
Therefore, those agents are going to use more tools than ever. This is actually an incredible time to be a software company. But the software has to be presented to the agent in a way that the agent can use it. This is a break big breakthrough and in fact what we have done, as you know, what Nvidia's treasure is is all of our CUDA libraries. I call them CUDA-X libraries. This is Nvidia's treasure.
## 00:15:52
**中文**
今天，我们现在能够将这些 CUDA-X 库呈现给智能体，他们甚至可以比人类更有效地使用它。因此，对于 CUDA-X 库来说，这是一个美好的时光。我们来看一下。 >> 20 年前，我们构建了 CUDA，一种用于加速计算的单一架构。我们重新发明了[音乐]计算。一千个 CUDA-X 库可帮助开发人员在科学和工程的各个领域取得突破。
**English / Original**
Today, we're able to now pre- present these CUDA-X libraries to agents who can use it much more effectively than even hu- humans. And so, this is a wonderful time for CUDA-X libraries. Let's take a look. >> 20 years ago, we built CUDA, a single architecture for accelerated computing. We reinvented [music] computing. A thousand CUDA-X libraries help developers make breakthroughs in every field of science and engineering.
## 00:16:26
**中文**
CUDA-X 库就是智能体的工具：[音乐] cuLitho 用于计算光刻，cuOpt 用于决策优化，cuDSS 用于直接稀疏求解，AI-Q 用于跨结构化与非结构化文档开展深度研究，Aerial 用于 AI RAN，Warp 用于可微分物理，Parabricks 用于基因组学。[音乐] 它们的基础都是算法，而算法是美丽的。
**English / Original**
CUDA-X libraries are tools [music] for agents. cuLitho for computational lithography. cuOpt for decision optimization. cuDSS for direct sparse solvers. AIQ for deep [music] research across structured and unstructured documents. Ariel for AI ran. Warp for differentiable physics. Parabricks [music] for genomics. At their foundation are algorithms and they are beautiful.
## 00:17:14
**中文**
>> [音乐] [音乐] [音乐] [音乐]
**English / Original**
>> [music] [music] [music] [music]
## 00:18:11
**中文**
[音乐] [音乐] [音乐] [音乐]
**English / Original**
[music] [music] [music] [music]
## 00:18:57
**中文**
[音乐] [音乐] [音乐] [音乐] [音乐] [音乐] [音乐]
**English / Original**
[music] [music] [music] [music] [music] [music] [music]
## 00:19:44
**中文**
[音乐][音乐][掌声]>>为数学鼓掌。数学是美丽的。软件的计算模式将会改变。事实上，让我们回到这一点。这是智能体。它是终极的分解和分布式计算的计算模型。
**English / Original**
[music] [music] [applause] >> A round of applause for math. Math is beautiful. The computing pattern the computing pattern of software is going to change. In fact, let's come back to this. This is the agent. It is the ultimate disaggregated and distributed computing computing model.
## 00:20:25
**中文**
为了处理一个智能体，许多不同的计算机会被调动起来。智能体由模型、编排框架、工具与技能以及运行时组成，这些部分分布在数据中心的不同位置运行。可以把模型看作大脑，把编排框架看作身体；工具则在运行时中工作，运行时就像一间工坊。
**English / Original**
So many different computers are going to be activated in order to process this agent. The agent consists of model, harness, tools and skills, and a runtime. All of that is running at different places in a data center. You can think of the model as the brain. The harness as the body. The tools that it uses working in a runtime, think of it as a workshop.
## 00:21:02
**中文**
这是一个在车间里使用工具工作的工人。当然，这是在非常大规模的情况下进行的。这些步骤中的每一步都在计算机的不同部分运行。你可以看到大型语言模型正在思考。情境处理、观察、理解环境、推理、提出计划并按照计划采取行动。每次发生这种情况，整个机架的 Grace Blackwell NVLink 72 都会被激活。它用大语言模型来思考。
**English / Original**
So this is a person a worker working with tools in a workshop. Of course, this is being done at extraordinarily large scales. And each one of those steps are running in a different part of the computer. And you could see the large language model is thinking. Context processing, observing, understanding the environment, reasoning, coming up with a plan, and acting on the plan. Every single time that happens, an entire rack of Grace Blackwell NVLink 72 is activated. It's thinking with the large language model.
## 00:21:41
**中文**
每当使用工具时，都会使用 CPU 使用率。该工具可以是 C 编译器。它可能是Python。它可能是 JavaScript。或者它可能是加速计算。今天的智能体是相对简单的工具用户。明天，他们将成为非常熟练的工具用户，这就是为什么我向您展示的 CUDA-X 库将非常受智能体欢迎的原因。
**English / Original**
Whenever it uses a tool, a CPU use is used. That tool could be a C compiler. It could be Python. It could be JavaScript. Or it could be accelerated computing. Today's agents are relatively simple users of tools. Tomorrow, they're going to be very sophisticated users of tools, which is the reason why the CUDA-X libraries that I showed you are going to be incredibly popular with agents.
## 00:22:09
**中文**
他们解决了世界上已知的一些最重要的问题。我们所有的 CUDA-X 库现在都将配备人工智能可以学习如何使用的技能。因此，CUDA-X 库、一些技能，基本上是一本手册，AI 读完后就会说：“啊哈，这就是你使用它的方式。”智能体使用这些库的能力将是令人难以置信的。因此，这些工具在 CPU 和 GPU 以及大型语言模型上运行。
**English / Original**
They solve some of the most important problems the world knows. And all of our CUDA-X libraries are now now going to come with skills that the AI could learn how to use. So, the CUDA-X library, some skills, basically a manual, the AI reads it and go, "Aha, that's how you use it." The ability to use these libraries by agents are going to be incredible. And so, the tools run on CPUs and GPUs and large language models.
## 00:22:42
**中文**
安全编排框架运行在 CPU 和一种称为 DPU 的安全处理器——NVIDIA BlueField——之上。整体编排运行在 CPU 上：这就是完整的框架，由 CPU 调度所有工作。最困难的部分之一是记忆。工作记忆称为 KV cache，其中涉及记住什么、如何做上下文压缩，以及不仅如何压缩、还如何检索。
**English / Original**
The security harness runs on CPUs and a security processor called a DPU, NVIDIA's BlueField. The orchestration of all this runs on a CPU. This is the entire harness, and the CPU's orchestrating all of the work. One of the hardest parts is memory. You could just imagine the working memory is called KV caching, what to remember, compaction, not just compression, but how to retrieve.
## 00:23:12
**中文**
您检索结构化数据吗？您检索非结构化数据吗？本体论是什么，所有这些不同数据与其自身的关系是什么？整个处理过程非常复杂。记忆系统，人工智能的记忆系统，将会引起存储系统的彻底变革。正如您所看到的，这种计算模型、这种计算模式、这种称为智能体的新应用程序的各个方面都与应用程序过去的运行方式有着根本的不同。
**English / Original**
Do you retrieve structured data? Do you retrieve unstructured data? What is the ontology, the relationship of all of these different data to itself? That entire processing is incredibly complicated. The memory system the memory system of AIs is going to cause the storage system to be completely revolutionized. As you could see, every aspect of this computing model, this computing pattern, this new application called an agent, is fundamentally different than the way that applications used to run.
## 00:23:48
**中文**
操作系统中二进制文件中的一大堆软件。这就是这种分解的原因，这种分布式的异构计算问题正是我们构建下一代 Vera Rubin 的原因。Vera Rubin 不是单一芯片。 Vera Rubin 不仅仅是 GPU。它从 GPU 开始，但 Vera Rubin 的表现令人难以置信。
**English / Original**
A whole bunch of software sitting inside a binary sitting inside an operating system. This is the reason this disaggregated, this distributed this heterogeneous computing problem is precisely the reason we built our next generation Vera Rubin. Vera Rubin is not one chip. Vera Rubin is not a GPU only. It starts with a GPU, but Vera Rubin is incredible.
## 00:24:21
**中文**
这整件事就是Vera Rubin。从头到尾。它有 GPU。 Vera Rubin NVLink 72。它是由 Vera CPU 编排的，我将向您介绍更多相关信息。存储系统具有革命性。 Vera 和 CX9 一起，我们的软件堆栈称为 DOCA。内置安全处理器，以便所有内容在静止、运动和使用时都被加密。
**English / Original**
This entire thing is Vera Rubin. From end to end. It has GPUs. Vera Rubin NVLink 72. It is orchestrated by Vera CPUs that I'm going to tell you more about. The storage systems revolutionary. Vera along with CX9, our software stack called Doka. The security processor that's inside so that everything is encrypted at rest, in motion, as well as in use.
## 00:24:56
**中文**
整个过程中的一切都是安全的，因为人工智能模型是如此珍贵。这就是整个系统服从机密计算的原因。这些系统中的每一个本身都是一场彻底的革命。 Vera Rubin 是我们公司历史上最雄心勃勃的事业。全公司 40,000 名工程师都在为 Vera Rubin 工作。更不用说你们所有人了。
**English / Original**
Everything across this is secure because the AI model is so precious. This is the reason why this entire system obeys confidential computing. Each one of these systems would be a complete revolution in itself. Vera Rubin is the most ambitious endeavor in the history of our company. The whole company worked on Vera Rubin across all 40,000 engineers. Not to mention all of you.
## 00:25:26
**中文**
你们所有人都参与了整个系统的创建。 Vera Rubin 确实是一个奇迹，它不仅仅是一个芯片，而是很多。嗯，甚至还不止于此。很久以前，NVIDIA 曾经是一家 GPU 公司。但多年来我们已经发展成为一家系统公司。您现在在这里寻找的是有史以来最复杂的系统，最复杂和最基础的系统。
**English / Original**
All of you participated in the creation of this entire system. Vera Rubin is really a miracle and it's not just one chip, it is so many. Well, it's even beyond that. A long time ago Nvidia used to be a GPU company. But over the years we've evolved to become a systems company. You're looking here now for the most complex systems, most complex and ground-up system ever designed.
## 00:25:56
**中文**
但最终，我们的客户、我们的合作伙伴不想购买计算机。他们想要建立AI 工厂，这就是NVIDIA真正开始再次转型的原因。你可以看到我们的很多技术现在都在整个基础设施规模上。我们的合作伙伴处于基础设施规模。发电机、冷却系统、电网提供商。
**English / Original**
But ultimately, our customers, our partners don't want to buy computers. They want to build AI factories, which is the reason why Nvidia has really started to transform ourselves yet again. You could see so much of our technology is now at the entire infrastructure scale. Our partners are at infrastructure scale. Power generators, cooling systems, the grid providers.
## 00:26:25
**中文**
现在有这么多工业公司成为我们生态系统的一部分，因为最终我们正在尝试构建一个像 GPU 一样的整个堆栈，就像我们构建 Grace Blackwell NVLink 72 时一样，就像现在一样，我们正在构建一个全栈系统，以便我们的客户可以构建令人惊叹的 AI 基础设施。我们来看一下。
**English / Original**
So many industrial companies are now part of our ecosystem because ultimately, we're trying to build an entire stack just like GPUs, just like when we were building Grace Blackwell NVLink 72, just like now, we are building a full-stack system so that our customers could build amazing AI infrastructure. Let's take a look.
## 00:26:52
**中文**
世界正在竞相建设AI 工厂，这是人类历史上最大的基础设施建设。AI 工厂非常复杂，每个[音乐]层、芯片、机架、网络、电源、冷却、>>[音乐]>>网格都必须从头到尾一起设计，因为计算就是收入。 >> [音乐] >> NVIDIA DSX 是蓝图，是以最高效率和盈利能力构建和运营 AI [音乐] 工厂的参考设计。
**English / Original**
The world is [music] racing to build AI factories, the largest infrastructure build-out in human history. AI factories are incredibly complex, every [music] layer, chip, rack, network, power, cooling, >> [music] >> grid, must be designed together from end to end because compute is revenues. >> [music] >> NVIDIA DSX is the blueprint, a reference design for building and operating AI [music] factories at maximum efficiency and profitability.
## 00:27:23
**中文**
它从 DSX Sim 开始。借助 DSX Sim [音乐] Omniverse 蓝图，合作伙伴可以在单机架 [音乐] 落地之前设计并验证 NVIDIA Vera Rubin AI 工厂。他们规划布局。模拟电源和冷却。 >> [音乐] >> 设计网络。验证每个集成。测试数字孪生中的每一个变化。
**English / Original**
It starts with DSX Sim. With the DSX Sim [music] Omniverse blueprint, partners design and validate an NVIDIA Vera Rubin AI factory before a single rack [music] lands. They plan the layout. Simulate the power and cooling. >> [music] >> Design the network. Validate every integration. Test every change in the digital twin.
## 00:27:50
**中文**
工厂通电。 DSX [音乐] 操作系统接管并配置、操作、监控和修复基础设施。 >> [音乐] >> 将已安装的系统转变为可信、多租户、有弹性、AI 就绪的[音乐]容量。如今的 AI 工厂的供电过剩率高达 40%。 DSX Max LPS [音乐] 让运营商能够在相同的功率预算内安全地部署更多 GPU。
**English / Original**
The factory powers on. DSX [music] OS takes over and provisions, operates, monitors, and remediates the infrastructure. >> [music] >> Turning the installed systems into trusted, multi-tenant, resilient, AI-ready [music] capacity. Today's AI factories over provision power by up to 40%. DSX Max LPS [music] lets operators safely deploy more GPUs inside the same power budget.
## 00:28:19
**中文**
增加数十亿[音乐]的年收入。 >> [音乐] >> 45° C 的突破性热液冷却使用更少的水和能源。更多的力量[音乐]将用于创收计算。极好的。动态功率分配将功率从[音乐]机架转移到另一个机架，回收滞留的瓦特，将其发送到正在工作的地方。
**English / Original**
Adding billions [music] in annual revenue. >> [music] >> Breakthrough hot liquid cooling at 45° C uses less water and energy. More power [music] going to revenue-generating compute. Incredible. Dynamic power allocation steers power from [music] rack to rack, recovering stranded watts, sending them where work is happening.
## 00:28:44
**中文**
机架内功率平滑可平滑峰值电流尖峰和功率浪涌。 [音乐] 在整个工厂中，AI 智能体团队与 DSX Max LPS 合作，不断[音乐]协调以平衡冷却和电源，以满足工作负载需求。 DSX AI 工厂是灵活的能源资产，可以与电网合作[音乐]运行。
**English / Original**
In-rack power smoothing flattens peak current spikes and power surges. [music] Throughout the factory, teams of AI agents work with DSX Max LPS, continuously [music] coordinating to balance cooling and power to meet workload demand. DSX AI factories are flexible energy assets that operate cooperatively [music] with the grid.
## 00:29:09
**中文**
DSX Flex 读取实时电网信号，并在电网需要救援时动态调整工厂功率 [音乐]。 100吉瓦的AI 工厂[音乐]将在本世纪末上线。 NVIDIA DSX AI 工厂[音乐]以最高效率运行，生产成本最低的token，并使电网更强大。 >> [掌声] >> 我向你们展示了过去的生态系统幻灯片。
**English / Original**
DSX Flex reads real-time grid signals and dynamically adjusts factory power [music] when the grid needs relief. A 100 gigawatts of AI factories [music] will come online before the end of the decade. Nvidia DSX AI factories [music] run at highest efficiency, produce the lowest cost tokens, and make the grid stronger. >> [applause] >> I've shown you ecosystem slides of the past.
## 00:29:45
**中文**
NVIDIA 的计算层和软件以及软件和计算堆栈被集成到服务于终端市场的其他人的平台、第三方平台和库中。那是一个计算生态系统。这是一个AI 工厂生态系统。这是你们所有人的下游。我的上游是你们大家，我们的下游是这个生态系统。
**English / Original**
Where Nvidia's computing layers and software and software and computing stacks are integrated into other people's platforms, third-party platforms and libraries that serves end markets. That was a computing ecosystem. This is an AI factory ecosystem. This is way downstream of all of you. Upstream of me is all of you and downstream of us is this ecosystem.
## 00:30:12
**中文**
NVIDIA 最终不仅是在造 GPU，也不仅是在造系统；我们是在帮助客户建设极其复杂的 AI 工厂和 AI 基础设施。这些设施达到 1 吉瓦规模时，投资起初约为 200 亿至 300 亿美元，现在是每吉瓦 500 亿至 600 亿美元，很快会达到每吉瓦 800 亿至 1,000 亿美元。
**English / Original**
Because Nvidia ultimately is not just building a GPU, not just building a system, we're helping customers build these AI factories, these AI infrastructure that is so immensely complex. Each one of these at 1 gigawatt level started at 30 20 30 billion dollars. It is at 50 60 billion dollars and soon it will be 80 100 billion dollars per gigawatt.
## 00:30:42
**中文**
1000亿美元建AI 工厂。它必须第一次就能起作用，并且必须立即起作用。资金成本令人难以置信。其复杂性令人难以置信。所以，正如你所看到的，我们过去常常在计算机内部设计芯片。然后我们在计算机内部模拟了一个系统。今天，你刚才看到了，一切都是在 Omniverse 中构建的。
**English / Original**
100 billion dollars into an AI factory. It must work the first time and it must work right away. The cost of capital is incredible. The complexity is incredible. So, as you see, we used to design a chip inside a computer. And then we simulated a system inside a computer. Today, you saw just now, everything was built in Omniverse.
## 00:31:11
**中文**
我和你们大家一起在 Omniverse 工作了很长时间。这就是梦想成真。因此，早在我们建立第一个突破口并投入资金之前，我们就可以在数字框架内、数字模拟器内、数字世界中构建世界想要构建的巨大系统。这就是我们的生态系统，我们称之为 DSX。
**English / Original**
I've been working with Omniverse with all of you for a long time. This was the dream come true. So, that we can build these gigantic systems as large as the world wants to build inside a digital framework, inside a digital simulator, in a digital world long before we build the first break ground and put our money to work. So, this is our ecosystem, our we call it DSX.
## 00:31:38
**中文**
RTX 对应我们的 GPU，DGX 对应我们的系统，而 DSX 对应基础设施。正因为我们在系统与软件等完整技术栈上所做的工作，我们才能与小公司合作，帮助它们成为世界一流的 AI 云。我接下来展示的企业大多成立不久；如今 CoreWeave 市值已达到数百亿美元，仍在飞速增长。最近我们也与 Nebius 展开合作。
**English / Original**
RTX is for our GPU. DGX is for our systems. And now DSX, basically, infrastructure. Because of the work that we do here across this entire stack, including our systems and software, it's the reason why we could work with small companies and enable them to be world-class AI clouds. Every one of these I'm about to show you are small companies just recently. And now CoreWeave is worth 50, 60, 70 billion dollars and growing incredibly fast. Recently, we worked with Nebius.
## 00:32:11
**中文**
同样，它们都在飞速增长，每个云平台都有出色的客户：软件编程公司 Cursor、从事图像生成的 Black Forest Labs、研发世界基础模型的 World Labs、领先的金融服务公司 Revolut，以及 Shopify。另一家云平台的客户还包括 BT 和 Google。
**English / Original**
And again, they're growing incredibly fast. Each one of these clouds have incredible customers. Cursor, the software coding company. Black Mountain Labs, image generation, World Labs, World Foundation model, Revolut, the leading financial services AI company, and Shopify. Here's another one. There's this Enscape, and their customers are British Telecom, Google.
## 00:32:38
**中文**
Google 正在使用我们的一个 AI 云；Thinking Machines 这家前沿实验室也是客户，我们对此非常兴奋。这里是韩国的 Naver Cloud，服务韩国银行、现代汽车等众多优秀企业；这里是印度的 Yotta；还有总部在新加坡、正在澳大利亚建设设施的 Together AI，以及 AI Singapore。
**English / Original**
Google is using one of our AI clouds, Thinking Machines, a Frontier Labs company. We're super exciting. Here's Naver Cloud in Korea. Bank of Korea, Hyundai, so many incredible companies. Here's one in in India, Yoda. Incredible companies. Here's one uh based in Singapore, building in Australia, Together AI. AI Singapore.
## 00:33:07
**中文**
这是印度尼西亚的一处。这些公司中的每一家都在为区域以及全球客户提供服务。人工智能将无处不在。每个公司都将受到它的推动。每个地区都会建设它。 Indosat 位于印度尼西亚。在台湾，GMI。在台湾，GMI。鼓掌就可以了。 >> [掌声] >> 所以，令人难以置信的令人难以置信的呃令人难以置信的公司，令人难以置信的机会，但它们都需要一些东西。的
**English / Original**
This is one in Indonesia. Each one of these companies, each one of these companies are serving regional as well as global customers. AI is going to run everywhere. Every company will be powered by it. Every region will build it. Indosat here in in Indonesia. Here in Taiwan, GMI. Here in Taiwan, GMI. It's okay to clap. >> [applause] >> So, incredible incredible uh incredible companies, incredible opportunity, but all of them need several things. Of
## 00:33:49
**中文**
当然，他们需要计算堆栈。
**English / Original**
course, they need the computing stack.
## 00:33:51
**中文**
这整个堆栈的底层，正是 NVIDIA 出名的原因。我们所有的硬件、软件和库，以及我们与世界第三方开发者生态系统的连接，使任何人都可以建立AI 云。然而，现在的AI 云是如此复杂。这是软件版本。这是计算机科学版本。
**English / Original**
This entire stack underneath, this is what made NVIDIA famous. All of our hardware and software and libraries, our connection into the world's ecosystem of third-party developers makes it possible for anyone to stand up an AI cloud. However, the AI cloud is so complex now. This is the software version. This is the computer science version.
## 00:34:16
**中文**
我先前展示的是资金与资产层面的版本：它是一座巨型工厂。只有计算栈还不够，这正是 NVIDIA 成为 AI 基础设施公司的原因。把这件事做好，真正擅长帮助客户建设和部署 AI 工厂，至关重要。原因如下。
**English / Original**
The money version, the asset version is what I showed you earlier. It's a giant factory. Having this ability alone is not enough, which is the reason why NVIDIA has become an AI infrastructure company. Now, doing this well and becoming incredibly good at at helping customers build AI factories and deploying AI factories is incredibly important. And the reason for that is this.
## 00:34:46
**中文**
现在计算就是收入。计算就是利润。没有收入和利润就是损失。因此，认识到这是AI 基础设施上线的一个例子非常重要。可能需要很长时间才能上线。这可能需要一段时间。它的吞吐量可能很高。它可能很低。
**English / Original**
Compute is revenue now. Compute is profit. The absence of revenues and profit is loss. And so, it's really important to realize that this is when this is an example of an AI infrastructure coming online. It could take It could be coming online quickly. It could take a while. Its throughput could be high. It could be low.
## 00:35:14
**中文**
它的弹性和可靠性可能是好是坏。它的使用寿命可长可短。因为这代表 50、60 到 1000 亿美元，所以这条曲线非常重要。这就是 NVIDIA 成为与我们合作的绝佳合作伙伴的原因，因为我们拥有完全集成的能力，我们不仅仅是制作了一张 PowerPoint 幻灯片。
**English / Original**
Its resilience and reliability could be good or bad. And its lifetime of usefulness could be long or short. Because this represents 50, 60, going to 100 billion dollars, this curve matters greatly. Which is the reason why NVIDIA is such a great partner working with us because of our fully integrated capability, we didn't just come up with a PowerPoint slide.
## 00:35:45
**中文**
我们创建了整个基础设施。我们将一切连接在一起。我们自己建造了数十亿个设备，以确保一切正常。因此，我们的第一个token的时间、我们的第一个token的时间、我们的首次推理时间、我们的训练时间都快得多。其次，因为我们的每瓦吞吐量、每瓦token绝对是世界一流的。
**English / Original**
We created the entire infrastructure. We connected everything together. We built out billions and billions of it ourselves to make sure that everything works well. As a result of that, our time our time to first token, our time to first token, our time to first inference our time to training turned on is much faster. Second, because our throughput per watt, our tokens per watt is utterly world-class.
## 00:36:22
**中文**
原因是我们整合了一切。我们从头开始设计一切。我们模拟整个系统并使用极端的协同设计。就像我刚才向您展示的 Vera Rubin 机架一样，一切的设计都是为了实现这种令人难以置信的吞吐量。如果你的数据中心、如果你的工厂有1吉瓦，就不会再有更多了。
**English / Original**
And the reason for that is because we integrate everything. We design everything from the ground up. We simulate the entire system and we use extreme co-design. Just like I showed you just now with the Vera Rubin rack, everything was designed in order to deliver on this incredible throughput. If your data center, if your factory has 1 gigawatt, it will not have more.
## 00:36:49
**中文**
1吉瓦是指1吉瓦。这就是你能做的所有发电。如果你有 1 吉瓦的电力，那么每瓦的吞吐量就是收入，因为每个token都是有利可图的。每个token都是收入。这就是未来。计算就是收入。每瓦性能就是您的收入。仅仅因为芯片更便宜而选择错误的架构并不能带来任何好处。
**English / Original**
1 gigawatt means 1 gigawatt. That's all the power generation you could do. If you have 1 gigawatt of power, then throughput per watt is revenues because every token is profitable. Every token is revenues. This is the future. Compute is revenues. Performance per watt is your revenues. Choosing the wrong architecture just because the chips are cheaper doesn't translate.
## 00:37:21
**中文**
没有道理。您需要确保您的每瓦收入越多，您购买的越多，您赚的就越多。所以每瓦token最后是最后哦，第二。第三是可靠性。如果您有机会参观这些数据中心，您会发现这里有如此多的移动部件、数百万根电缆。所有这些计算机和谐、可靠地工作的能力极低。这是极其困难的。我们现在已经进行了很长时间的大规模运营。这种经历很重要。
**English / Original**
Doesn't make sense. You need to make sure that your revenues per watt the more you buy, the more you make. And so tokens per watt and then lastly very last Oh, second. Third is reliability. If you ever get a chance to see these data centers, there are so many moving parts, millions of cables. The ability for all of those computers to work harmoniously, reliably, is extremely low. It is just extremely difficult. We have now been operating very large scale for a very long time. That experience matters.
## 00:38:02
**中文**
这种差异意味着中断之间的时间极其重要。最后，这非常困难。这些系统的生命周期，软件的生命周期一直在变化。四年前，也就是霍珀的时代。人工智能已经完全改变了。六年前，这是安培的时间框架，AI彻底改变了。
**English / Original**
That difference mean time between interrupts extremely important. And then lastly this is very hard. The lifetime of these systems, the lifetime of these systems the software is changing all the time. 4 years ago which is in the time of Hopper. AI has completely changed. Six years ago, this is the time frame of Ampere, AI has completely changed.
## 00:38:33
**中文**
我们开始谈论 CNN。我们在这里，然后我们谈论了Transformer，然后我们谈论了专家的混合。现在我们谈论智能体系统。每一代、每几个月，软件行业都会推出新技术。如果你的架构不灵活，如果你的生态系统不丰富，那么这个曲线不可能很长。
**English / Original**
We started out talking about CNNs. Here we are, then we talked about Transformers, and then we talked about mixture of experts. Now we're talking about agentic systems. Every single generation, every single few months, the software industry is coming up with new technology. If your architecture is not flexible, if your ecosystem is not rich, then this curve cannot be long.
## 00:39:03
**中文**
您无法预测您的系统可以持续多久。我可以。 NVIDIA 系统遍布全球。软件开发人员从 NVIDIA CUDA 开始，因此根据定义，其生命周期、生态系统和有用资产将会更长。差异本质上是成本。你可以将其视为收入，但收入的另一面是成本。
**English / Original**
You cannot predict how long your system can last. I can. Nvidia systems is all over the world. Software developers start with Nvidia CUDA, and by definition, therefore, the life, the ecosystem, the useful asset is going to be much longer. The difference is essentially cost. You could think of it as revenues, but the other side of revenues is cost.
## 00:39:30
**中文**
如果资产的寿命较长，则 TCO 较低。这就是区别。这就是计算时的样子。你买的越多，赚的就越多。 >> [掌声] >> 现在，你们所有人都和我一起经历了这一点。是不是这样？你们所有的需求，你们的工厂都这么努力，全台湾你们的人都这么努力，因为
**English / Original**
If the life of the asset is long, the TCO is low. This is the difference. This is what it looks like when compute The more you buy, the more you make. >> [applause] >> Now, all of you are experiencing this with me. Isn't that right? All of your demand, your factories are working so hard, your people are working so hard all across Taiwan, because
## 00:40:09
**中文**
每个人都想赚钱。
**English / Original**
everybody wants to make money.
## 00:40:12
**中文**
他们意识到人工智能，有用的人工智能就在这里。有利可图的人工智能就在这里。计算需求非常高，而计算需求是制约因素。所以，让我们超级超级努力地工作，帮助世界各地建立AI 工厂。这就是为什么它如此重要。我太高兴了。我就站在你面前。Vera Rubin 已全面投入生产。
**English / Original**
They realize that AI, useful AI is here. Profitable AI is here. Compute demand is incredibly high and compute demand is the constraint. And so, let's go work super super hard and help the world stand up AI factories everywhere. This is why it's so important. I'm so happy. Here I am standing in front of you. Vera Rubin is in full production.
## 00:40:45
**中文**
>> [掌声] >> Vera Rubin 正在全力制作。我们为 Vera Rubin 创建的供应链是 Grace Blackwell 的两倍。不是啊，这太不可思议了。过去需要 2 个小时才能组装一个 Grace Blackwell 机架，现在只需 5 分钟。因此，不仅容量更高，吞吐量也更快。
**English / Original**
>> [applause] >> Vera Rubin is in full production. It the the supply chain we created for Vera Rubin is twice as large as Grace Blackwell. Not Yeah, it's incredible. And And what used to take 2 hours to assemble one Grace Blackwell rack now only takes 5 minutes. So, not only is the capacity higher, the throughput is a lot faster.
## 00:41:17
**中文**
我们需要一切来支持需求。这个生态系统非同寻常。数百万平方英尺的面积已投入使用，以支持Grace Blackwell (Grace Blackwell)，并立即做好准备，Vera Rubin 正在全力以赴。我要感谢你们所有人。Vera Rubin 现已全面投入生产。谢谢。 >>[掌声]>>让我们来看看。 >> 大型语言模型生成[音乐]答案。
**English / Original**
And we need it all to support the demand. This ecosystem is extraordinary. Millions of square feet has been put online to support Grace Blackwell and preparing now, ramping up now, Vera Rubin. I want to thank all of you. Vera Rubin is now in full production. Thank you. >> [applause] >> Let's take a look. >> Large language models generate [music] answers.
## 00:41:49
**中文**
现在AI 智能体可以工作了。但处理智能体[音乐]人工智能是一个完全不同的问题。智能体观察、推理、计划、使用工具。他们管理大量的[音乐]环境，兼顾工作记忆和长期记忆。他们根据需要启动子智能体、专家[音乐]。 NVIDIA Vera Rubin 是一款多机架 Pod 规模系统 >> [音乐] >> 专为处理 gentic AI 而构建，现已全面投入生产。整个供应链的制造、自动化和编排[音乐]，这是一个有目共睹的奇迹。
**English / Original**
Now AI agents can do work. But processing agentic [music] AI is a whole different kind of problem. Agents observe, reason, plan, use tools. They manage massive [music] context, juggling working memory and long-term memory. They spin up sub agents, specialists [music] on demand. Nvidia Vera Rubin is a multi-rack pod scale system >> [music] >> built to process a gentic AI and is now in full production. The manufacturing, automation, and orchestration [music] across the supply chain, a miracle to witness.
## 00:42:23
**中文**
当我们推出第一台人工智能超级计算机 NVIDIA DGX-1 时，我们的旅程就开始了。 [音乐] 在接下来的十年里，我们将每一个芯片和系统都推向了极限。从 Pascal [音乐] 和第一台 NVLink 到第一台机架级 AI 超级计算机 Grace Blackwell。现在，Vera Rubin 是第一台为智能体时代打造的多机架 Pod [音乐] 规模超级计算机。
**English / Original**
Our journey started when we launched the first AI supercomputer, Nvidia DGX-1. [music] Over the next decade, we pushed every chip and system to the limit. From Pascal [music] and the first NVLink to Grace Blackwell, the first rack-scale AI supercomputer. And now, Vera Rubin, the first multi-rack pod [music] scale supercomputer built for the agentic age.
## 00:42:46
**中文**
它始于台积电。构成 Vera Rubin 的七个新芯片通过数百个处理步骤形成了[音乐]。 3纳米工艺，CoWoS R和CoWoS [音乐] L封装，来自Micron、SK Hynix和Samsung的HBM4内存。 Vera Rubin 在一块板上使用超过 18,000 个组件计算 6 万亿个晶体管[音乐]。
**English / Original**
It starts at TSMC. The seven new chips that make up Vera Rubin take shape [music] through hundreds of processing steps. 3 nanometer process, CoWoS R and CoWoS [music] L packaging, HBM4 memory from Micron, SK Hynix, and Samsung. The Vera Rubin compute 6 trillion transistors [music] with over 18,000 components on one board.
## 00:43:10
**中文**
Vera Rubin NVL72 负责思考、提示和上下文理解、[音乐]推理和规划。接下来，新的模块化计算托盘、采用新 PCB 中板设计的精简 [音乐]、超级芯片、ConnectX-9 超级 NIC 和 BlueField-4 [音乐] DPU 全部配合到位，无需电缆，可实现 AI 工厂规模的弹性。 18 个计算[音乐]托盘、9 个热插拔 NVLink 开关托盘、新型高效歧管、承载超过 5,000 安培电流的液冷[音乐]母线，相当于全加速时的 20 辆电动汽车。
**English / Original**
Vera Rubin NVL72 does the thinking, prompt and context understanding, [music] reasoning, and planning. Next, a new modular compute tray, streamlined [music] with a new PCB midplane design, superchips, ConnectX-9 super NICs, and BlueField-4 [music] DPUs all mate in place with no cables for resiliency at AI factory scale. 18 compute [music] trays, nine hot-swappable NVLink switch trays, new high-efficiency manifolds, liquid-cooled [music] bus bars carrying over 5,000 amps, the equivalent of 20 electric cars at full acceleration.
## 00:43:49
**中文**
130 万个组件 >> [音乐] >> 共同构成了第三代 MGX 机架设计。祝贺 Microsoft 推出 Vera Rubin NVL72 工程机架。也祝贺 Dell 和 CoreWeave 推出了 Vera Rubin NVL72 工程机架。然后是 Vera CPU 机架。单个液冷机架中有 256 个 [音乐] CPU。
**English / Original**
Together, 1.3 million components >> [music] >> form this third-generation MGX rack design. Congratulations to Microsoft for their operational Vera Rubin NVL72 engineering rack. Congratulations to Dell and CoreWeave as well for standing up their Vera Rubin NVL72 engineering rack. Then, the Vera CPU rack. 256 [music] CPUs in a single liquid-cooled rack.
## 00:44:15
**中文**
编排模型。洗牌记忆。 >> [音乐] >> 启动工具。在富士康和广达，Groq 3 LPX 已初具规模。 256 Grok [音乐] 16 个托盘上的 3 个 LPU。每秒 40 PB 的 SRAM 带宽，可实现超低延迟。 MVL 72 [音乐] 以最高吞吐量生成token，而 Groq LPX 以最低延迟生成token。
**English / Original**
Orchestrating the models. Shuffling memory. >> [music] >> Launching tools. At Foxconn and Quanta, Grok 3 LPX takes shape. 256 Grok [music] 3 LPUs across 16 trays. 40 petabytes per second of SRAM bandwidth for ultra-low latency. While MVL 72 [music] generates tokens at the highest throughput, Grok LPX generates them at the lowest latency.
## 00:44:43
**中文**
Vera BlueField 4 STX，[音乐] AI 保存记忆的地方。 BlueField 4 加速存储处理。连接内存、存储和硅内安全性。以及 NVIDIA Spectrum-X [音乐] 以太网光子学。世界上第一台具有 200 Gb [音乐] 共同封装光学器件的以太网交换机。台积电的蒸煮工艺、芯片级封装以及磷化铟上的超高功率[音乐]激光二极管。
**English / Original**
Vera BlueField 4 STX, [music] where AI keeps its memory. Storage processing accelerated by BlueField 4. Connecting memory, storage, and in-silicon security. And NVIDIA Spectrum-X [music] Ethernet photonics. The world's first Ethernet switch with 200 gigabit [music] co-packaged optics. TSMCs cook process, chip-scale packaging, and ultra-high-powered [music] laser diodes on indium phosphide.
## 00:45:13
**中文**
Vera Rubin.五个连接的[音乐]机架规模系统。AI 智能体的超级计算机。全台湾150家供应链[音乐]合作伙伴。数百万平方英尺的工厂车间。数百个[音乐]站点、芯片、封装、系统和数据中心被推向了尺寸、功率和规模的极限。这就是我们所说的极端协同设计。 [音乐]我们和台湾就是这么做的。我们共同重塑了人工智能时代的计算。
**English / Original**
Vera Rubin. Five connected [music] rack-scale systems. A supercomputer for AI agents. 150 supply chain [music] partners across Taiwan. Millions of square feet of factory floor. Hundreds [music] of sites, chips, packages, systems, and data centers pushed to the limits of size, power, and scale. This is what we call extreme co-design. [music] We did this with Taiwan. Together, we reinvented computing for the age of AI.
## 00:45:42
**中文**
台湾从一开始就与我们同在，今天我们将 Vera Rubin 带到了世界。谢谢你，台湾。 >> [音乐] [掌声] >> 女士们先生们，Vera Rubin。 >> [掌声] >> Vera Rubin 不仅仅是为人工智能而生。 AI Vera Rubin 的诞生不仅仅是为了运行 AI。 Vera Rubin 是为运行智能体而生的。这是一个智能体系统。想象一下复杂性，这就是智能体成为计算机科学最后突破的原因。它有
**English / Original**
Taiwan was [music] with us at the beginning and here today as we bring Vera Rubin to the world. Thank you, Taiwan. >> [music] [applause] >> Ladies and gentlemen, Vera Rubin. >> [applause] >> Vera Rubin was not just built for AI. AI Vera Rubin was not built just to run AI. Vera Rubin was built to run agents. This is an agentic system. Imagine the complexity, which is the reason why agents is the last computer science breakthrough. It has
## 00:46:24
**中文**
智能体花了很多年才意识到其潜力并变得有用。按理说，运行它的计算机是世界上最先进的。这是Vera Rubin。我们来看一下。我们可以把Vera Rubin带出来吗？ >> [掌声] >> 工程部，我们有机架、系统吗？
**English / Original**
taken this many years for agents to realize its potential and become useful. It stands to reason that the computer that runs it is the most advanced in the world. This is Vera Rubin. Let's take a look. Can we bring out Vera Rubin, please? >> [applause] >> Engineering, do we have the Do we have the racks, the systems?
## 00:47:07
**中文**
看起来很重。
**English / Original**
It looks heavy.
## 00:47:10
**中文**
这是这是Vera Rubin。 Vera Rubin NVLink 72。这是 Groq LPX。在下一届 GTC 上，我将与您讨论更多相关内容。今天，我们有很多话要和大家聊。这是Vera的CPU机架。 256 个 CPU，全液冷。让我稍后告诉你关于Vera的事。这是 Vera BlueField 存储处理系统和安全系统。当然，这是我们的 Mellanox 网络，世界上第一个 CPO。
**English / Original**
This is This is Vera Rubin. Vera Rubin NVLink 72. This is the Grock LPX. At the next GTC, I'm going to talk to you about a lot more of this. Today, we have so much to talk to you about. This is Vera's CPU rack. 256 CPUs, all liquid cooled. Let me tell you about Vera in just a moment. This is the Vera BlueField storage processing system and also security system. And of course, this is our Mellanox networking, the world's first CPO.
## 00:47:49
**中文**
这是Vera Rubin。令人难以置信的技术全部汇集在一起。现在，当我们构建 Hopper 时，如您所知，我们构建 Hopper 是为了进行预训练。预训练是最重要的应用，也是我们当时做的最重要的工作量。然后，当我们研究 Grace Hopper Blackwell 时，每个人都说：“黄仁勋，你知道，NVIDIA 非常擅长预训练。
**English / Original**
This is Vera Rubin. Incredible technology all coming together. Now, when we built When we built Hopper, we built Hopper, as you know, for pre-training. Pre-training was the most important application, the most important workload we were working on at the time. Then when we worked on Grace Hopper Blackwell everybody said, "Jensen, you know, Nvidia is really good at pre-training.
## 00:48:13
**中文**
推理是如此容易。”你还记得吗？人们过去常说：“推理是如此容易。我们也可以做到。”但如您所知，推理等于金钱。而且模型、MOE 非常复杂。要在令人难以置信的高响应时间、快速交互性和高吞吐量的同时做到这一点非常困难。这就是我们创建 NVLink 72 的原因。
**English / Original**
Inference is so easy." Do you remember that? People used to say, "Inference is so easy. We could do that, too." But as you know, inference equals money. And the models, MOEs are so complicated. And to do it at incredibly high response time fast interactivity, and high throughput at the same time is incredibly hard. Which is the reason why we created NVLink 72.
## 00:48:41
**中文**
如今，NVIDIA的token成本是全球最低的。不是 10%，而是 X 个因素，几个数量级。这一切都是因为我们进行了极端的协同设计。这一切都是因为我们了解计算模型、推理的计算模式，并且我们能够创建 NVLink 72。现在有了 Vera Rubin，它就超越了推理。现在它是在智能体智能体系统中的推理。
**English / Original**
Today, Nvidia's token cost is the lowest in the world. Not by 10%, by X factors, orders of magnitude. All because we did extreme co-design. All because we understood the computing model, the computing pattern of inference and we were able to create NVLink 72. Now with Vera Rubin it is beyond inference. It is now inference in an agentic agentic system.
## 00:49:12
**中文**
这是Vera Rubin。没有电缆、软管、风扇。上次我向你们展示这个时，我们发现到处都有电缆。这些电缆看起来棒极了。但现在中间有一块PCB连接两侧。以前需要2个小时的事情现在只需5分钟。Vera Rubin 的可靠性和韧性将是超乎寻常的。
**English / Original**
This is Vera Rubin. No cables no hoses, no fans. What used to take the last time when I showed this to you, we had cables everywhere. The cables were amazing to look at. But now there's a PCB in the middle which connects both sides. What used to take 2 hours now takes 5 minutes. The reliability and the resilience of Vera Rubin is going to be off the charts.
## 00:49:45
**中文**
这是我们的 Vera CPU 托盘。迄今为止最先进的 CPU。我将在一秒钟内向您展示这一点。这是我们的储物托盘。两个 Vera CPU、四个 CX 9、数量惊人的软件。这是我们新的 LPX LPU 30，Groq 系统专为极低延迟推理而设计。吞吐量由 Vera Rubin 提供，并通过 NVLink 72 进行扩展。
**English / Original**
This is our Vera CPU tray. The most advanced CPUs that has ever been built. I'm going to show you that in just a second. And this is our storage tray. Two Vera CPUs, four CX 9, incredible amounts of software. This is our new LPX LPU 30, the Groq system designed for very low latency inference. The throughput is delivered by Vera Rubin and extended with NVLink 72.
## 00:50:23
**中文**
如果您想进一步扩展，您可以使用 Groq LPU。这里，我们有 Vera Rubin NVLink，开关托盘。这是中间的开关，这是革命性的。因为 Vera Rubins，因为 NVLink 72，以及我们创建和发明的 NVLink 交换机，这就是我们用于横向扩展的以太网交换机。
**English / Original**
If you want to extend that even further, you can have Groq LPUs. Here, we have the Vera Rubin NVLink, the switch tray. This is the switches in the middle, and this is revolutionary. Because of Vera Rubins, because of NVLink 72, and the NVLink switches that we created and invented, and this is our Ethernet switches for scale out.
## 00:50:52
**中文**
令人惊奇的是我们为 Grace Blackwell 引入了这两个系统。这两个系统是为 Grace Blackwell 创建的，如今，NVIDIA 是世界上最大的网络公司。我为网络团队感到非常自豪。这对于我们所做的一切来说都是一个令人难以置信的推动者。我现在要和你们谈谈我们将要参与的下一个主要行业。
**English / Original**
What's amazing is we introduced these two systems for Grace Blackwell. These two systems were created for Grace Blackwell, and today, NVIDIA is the largest networking company in the world. I'm so proud of the networking team. This is such an incredible enabler for everything that we do. I'm going to now talk to you about the next major industry we're going to be part of.
## 00:51:21
**中文**
谢谢。仁勋.谢谢。 >> [掌声] >> 黄仁勋。我认为有 2,000 人在背后推动着这一点。好吧，我们来谈谈CPU。Vera CPU。专为人工智能时代打造的 CPU。迄今为止所有的 CPU 都是为人类而设计的。
**English / Original**
Thank you. Jen-Hsun. Thank you. >> [applause] >> Jensen. I think there are 2,000 people back there pulling that. Okay, let's talk about CPUs. Vera CPUs. CPUs built for the age of AI. All of the CPUs until now were created for people.
## 00:52:00
**中文**
我们是用户。我们是用户。我们是租客。根据我们使用 CPU 的方式，我们生活在一个以秒为单位的世界中。我们在云中租用CPU的方式是，每一个CPU越多，您拥有的CPU核心越多，您可以租用的CPU就越多。旧 CPU 的经济性、旧 CPU 的用例以及旧 CPU 的经济性与智能体根本不同。
**English / Original**
We were the users. We were the users. We were the renters. The way we use CPUs, we live in a world counted by seconds. The way we rent CPUs in the cloud, each one of them more you can more CPU cores you have, the more you can rent. The economics of the old the use case of the old CPU and the economics of the old CPU fundamentally different than agents.
## 00:52:30
**中文**
智能体们已经不耐烦了。他们并不生活在一个以秒为单位的世界。他们生活在一个以纳秒为单位的世界。当它使用工具时，它希望响应时间尽可能快。当它访问数据库时，它必须尽快返回。智能体等待的每一刻都会阻止它进入下一步、下一步、下一步。
**English / Original**
Agents are impatient. They don't live in a world that is in seconds. They live in a world that's in nanoseconds. When it uses a tool, it wants the response time to be as fast as possible. When it access database, it has to come back as soon as possible. Every moment that the agent is waiting keeps it from going to the next step, the next step, the next step.
## 00:52:58
**中文**
至关重要的是，我们必须使 CPU 的延迟尽可能低，并尽可能具有交互性。因此，我们为人工智能时代创建了 Vera CPU。现在，在我们的系统中，它以三种不同的方式使用。第一种方式当然是Vera Rubin（Vera Rubin）的思考方式。在 Vera Rubin 机架内，已经有两个 CPU。如您所知，我们正在制造和销售数百万辆Vera Rubin 系统。我们已售出数百万张 Grace Blackwell。NVIDIA已经是其中之一
**English / Original**
It is vital that we make the CPUs as low latency as possible, as interactive as possible. So, we created Vera CPU for the age of AI. Now, inside our system, it's used for three different ways. The first way, of course, is Vera Rubin for thinking. And inside the Vera Rubin rack, there already two CPUs. As you know, we are building and selling millions of Vera Rubins. We have sold millions of Grace Blackwalls. Nvidia already is one
## 00:53:39
**中文**
全球最大的 CPU 制造商之一。
**English / Original**
of the largest CPU makers in the world.
## 00:53:42
**中文**
Vera Vera Rubin 机架中有两个 CPU。一种用于编排和管理 GPU、管理 KV 缓存、处理机架中运行的所有软件。我们还有用于安全和隔离的 Grace BlueField。 Vera 计算用于人工智能模型的利用、编排、工具使用、访问数据库。
**English / Original**
Vera In the Vera Rubin rack are two CPUs. One for orchestrating and managing the GPUs, managing the KV cache, dealing with all of the software that runs in the rack. We also have the Grace Bluefield that is used for security and isolation. The Vera compute is used for the harness, the orchestration of the AI models, tool use, accessing the database.
## 00:54:15
**中文**
数据服务器就在这里，Vera BlueField。最快的存储 最快的存储服务器，世界上最快的存储系统。这之所以如此重要，是因为智能体正在访问内存，访问内存的速度非常快。这些系统、存储服务器和CPU现在是数据中心最昂贵部分的关键路径。
**English / Original**
And the data servers are right here, Vera Bluefield. The fastest storage fastest storage servers, the fastest storage system the world has ever made. And the reason why this is so vital is because agents are accessing memory, accessing memory so incredibly fast. These systems, the storage server and the CPUs are now the critical path of the most expensive part of the data center.
## 00:54:46
**中文**
这是最贵的，这是有充分理由的。AI 工厂的经济学是token。token是在这里创建的。因此，您当然希望制造和生成尽可能多的token。这是你投入所有经济知识的地方。这一定不能成为障碍。所以Vera CPU对Vera在CPU架构上的压力很大，这也是我们从头开始打造全新架构的原因。世界上从未见过的CPU。
**English / Original**
This is the most expensive for a good reason. The economics the economics of the AI factory is tokens. And the tokens are created here. And so of course you want to manufacture and generate as many tokens as possible. This is where you put all of your economics. And this has to not be in the way. And so Vera CPU has great pressure on the Vera on the CPU architecture, which is the reason why we built a brand new architecture from the ground up. A CPU the world has never seen before.
## 00:55:24
**中文**
我们称它为 Vera：面向智能体的 CPU。过去的 CPU 是为人类打造的，而这颗 CPU 是为智能体打造的。需要记住四个要点。第一，Vera 必须拥有世界一流的 IPC（每时钟周期指令数），因为我们需要低延迟；我们看重的是处理时间和单线程性能，而不只是吞吐量。
**English / Original**
We call it Vera. This is CPU for agents. All the CPUs of the past we built for humans. This CPU is built for agents. Well, there four things to keep in mind. The four takeaways. The first takeaway is that the instructions per clock of Vera has to be incredibly good because we need the latency to be short. We need the processing time, single threaded performance, not throughput, single threaded performance has to be world-class, absolutely the best.
## 00:56:02
**中文**
单线程性能，这就是Vera的IPC、每时钟指令如此高的原因。这是世界上最高的。每个时钟读取、解码和执行 10 条指令。第一。第二。 CPU 传入和传出数据所需的带宽必须是世界一流的。第二件事是每个核心的带宽。
**English / Original**
Single threaded performance, which is the reason why the IPC, the instructions per clock of Vera is so high. It's the highest in the world. 10 instructions fetch, decoded, and executed per clock. Number one. Number two. The bandwidth necessary to move data in and out for the CPU has to be utterly world-class. The second thing is bandwidth per core.
## 00:56:30
**中文**
第三项是整体带宽。请记住，我之前说过，智能体系统从根本上是解耦且分布式的。一旦计算被解耦并分布到不同位置，网络就成为关键，因此我们必须尽可能快地在 CPU 核心之间、CPU 与存储之间、以及 CPU 与 GPU 之间移动数据。
**English / Original**
The third is just bandwidth, period. We're moving Remember, I said earlier, agentic systems is fundamentally disaggregated and distributed. Disaggregated and distributed. When computing is disaggregated and distributed, networking becomes the problem. Therefore, we have to move the data around as fast as possible between the CPU cores and between the CPU and the storage, the CPU and the GPU.
## 00:57:02
**中文**
系统周围和 CPU 核心内部的带宽必须完全是世界一流的。这是第一个经过很长时间构建的 CPU，其结构以光速连接所有 CPU 核心，实际上已达到理论极限。每秒 3.6 TB。没有小芯片税，没有芯片边界交叉，因为我们需要拥有一切，因为 CPU 内核以极高的带宽相互通信。它们不是按每个核心每个核心租用的。他们都在一起工作。
**English / Original**
The bandwidth around the system and inside the CPU core has to be utterly world-class. This is the first CPU that's been built a long time that is literally at theoretical limits with a fabric that connects all of the CPU cores that is speed of light. 3.6 terabytes per second. No chiplet tax, no chip boundary crossings because we need to have everything because the CPU cores are talking to each other with extremely high bandwidth. They're not rented core per core per core. They're all working together.
## 00:57:41
**中文**
Vera 的横截面带宽超乎想象。它是第一款采用 PCI Express Gen 6 的产品。它也是第一款采用每秒 1.2 TB 的 LPDDR5 的产品。外部最高性能 CPU 带宽的三倍，内部带宽的三倍。每个核心的带宽和带宽周期都是世界一流的。
**English / Original**
The cross-sectional bandwidth of Vera is off the charts. It's the first one to be PCI Express Gen 6. It is also the first one to have LPDDR5 with 1.2 terabytes per second. Three times two to three times the bandwidth of the highest performance CPUs on the outside three times the bandwidth on the inside. The bandwidth per core and the bandwidth period is world class.
## 00:58:11
**中文**
现在，请记住我之前向您展示了 CPU 核心的数量，CPU 的数量将会相当高。原因很简单。我们过去为人类创造了CPU。而人类，只是我们中的 10 亿。将会有数十亿的智能体。这些智能体将毫无耐心地使用 CPU，因为它们旁边的 GPU 的成本太高了。
**English / Original**
Now, remember I showed you earlier the number of CPU cores, the number of CPUs is going to be quite high. And the reason for that is very simple. We created CPUs for humans in the past. And humans, they're only 1 billion of us. There will be billions of agents. And these agents are going to be using the CPUs with very little patience because the cost of the GPU they sit next to is too high.
## 00:58:45
**中文**
因此太有价值、太珍贵了。因此，这些 CPU 不仅要具有高性能，而且还必须非常节能。这样我们就可以在工厂中塞入尽可能多的 CPU，而不会剥夺token生成的能力，我们知道这是我们赚钱的方式。 Vera 定义为每时钟指令数或每核单线程性能带宽、芯片周围和芯片内部的总带宽以及能效这四个属性。这绝对是世界
**English / Original**
And therefore too valuable, too precious. Therefore, these CPUs are going to be both performant but they also have to be extremely energy efficient. So that we can cram as much CPUs we can into the factory without taking away power from the token generation, which we know is how we make money. These four properties, instructions per clock or single-threaded performance bandwidth per core the total bandwidth around the chip and inside the chip and energy efficiency defines Vera. It is absolutely world
## 00:59:27
**中文**
世界一流。与性能最高的 x86 CPU 相比，它的优势非常显著；在真实的单线程工作负载中，它打破了纪录。CPU 性能提升 5% 已经很难得，能提升 10% 更是非凡，而这里展现的提升幅度前所未有。
**English / Original**
class. When you compare it to the highest performance x86, it is just off the charts. When you compare it in real single-threaded performance, real performance, it's off the charts. It is incredible to be able to deliver 5% improvement on CPUs. It is incredible to be able to deliver 10%. But this kind of performance speed up is just unheard of.
## 00:59:55
**中文**
这是 NVIDIA Vera。你怎么认为？ >>[掌声]>>让我们来看看。 >> 生成式人工智能改变了 CPU 的角色。现在，CPU 是指挥，GPU 是[音乐]管弦乐队。传统 CPU 是为不同时代而设计的，最大限度地提高了每个插槽的内核 [音乐] 数量。将它们分割、虚拟化、按小时出租。
**English / Original**
This is Nvidia Vera. What do you think? >> [applause] >> Let's take a look. >> A generative AI changes the role of the CPU. The CPU is now the conductor and the GPU is [music] the orchestra. Traditional CPUs were built for a different era, maximizing cores [music] per socket. Slice them up, virtualize, rent by the hour.
## 01:00:25
**中文**
在智能体时代，CPU 现在成为 GPU 利用率的瓶颈，直接影响[音乐]token吞吐量、延迟和用户体验。 NVIDIA [音乐] Vera 是专为智能体循环构建的 CPU，它将 NVIDIA 的定制数据中心 CPU 核心与可扩展的一致性结构 [音乐] 相结合，以实现性能核心和带宽的适当平衡，从而最大限度地提高 AI 工厂的输出。
**English / Original**
In the age of agents, the CPU is now a bottleneck to GPU utilization, directly affecting [music] token throughput, latency, and user experience. Nvidia [music] Vera is the CPU built for the agentic loop, combining Nvidia's custom data center CPU core with a scalable coherency fabric [music] for the right balance of performance cores and bandwidth to maximize AI factory output.
## 01:00:51
**中文**
[音乐] Vera 的核心是 NVIDIA Olympus 核心，专为现代数据中心工作负载、分支繁重的 [音乐] Python 运行时、工具调用和沙箱代码执行而构建。每个核心都针对吞吐量进行了调整。神经分支预测器 >> [音乐] >> 每个周期评估两个采用的分支，10 宽解码引擎每个周期带来更多工作。
**English / Original**
At the heart of [music] Vera is the Nvidia Olympus core, built for modern data center workloads, branch-heavy [music] Python runtimes, tool calls, and sandbox code execution. Each core is tuned for throughput. A neural branch predictor >> [music] >> evaluating two taken branches per cycle, a 10-wide decode engine brings in more work each cycle.
## 01:01:13
**中文**
>> [音乐] >> 大型乱序执行引擎让指令持续流动；先进的预取器配有新型图引擎，用来预测下一条数据路径。但只有数据正确且及时到达，快速核心才有意义。[音乐] Vera 是首款采用 LPDDR5X 内存的 CPU，[音乐] 能够同时纠正多项错误而不牺牲带宽。
**English / Original**
>> [music] >> A large out-of-order engine keeps instructions moving. Advanced prefetchers with a novel graph engine anticipating the next data path. But fast cores [music] only matter when data arrives correctly and on time. Vera is the first CPU [music] to use LPDDR5X memory while correcting multiple errors simultaneously [music] without compromising bandwidth.
## 01:01:36
**中文**
与 X86 相比，Vera 的峰值内存延迟降低了 40%，通过检索、分析和沙箱执行保持内核按时运行。 NVIDIA 的第二代可扩展[音乐]一致性结构将所有 88 个 Olympus 核心统一在一个整体网格上，并具有用于内存[音乐]和 IO 的单独芯片。内核不会跨小芯片分割，因此比传统 CPU 的内核间通信速度提高 50% [音乐]。
**English / Original**
Vera achieves 40% lower peak memory latency versus X86, keeping cores fed on time through retrieval, analytics, and sandbox execution. Nvidia's second-generation scalable [music] coherency fabric unifies all 88 Olympus cores on a monolithic mesh with separate dies for memory [music] and IO. Cores are not split across chiplets, enabling 50% [music] faster core-to-core communication than traditional CPUs.
## 01:02:06
**中文**
内存[音乐]一致的 NVLink 芯片到芯片将 GPU 直接连接到结构。除了 GPU 之外，[音乐] NVLink 芯片到芯片还可以将 Vera 扩展到多个插槽，从而在 CPU 之间实现巨大的带宽。 >> [音乐] >> Vera 提供的智能体沙箱性能是 X86 CPU 的 1.8 倍。独立的 Vera 机架运行智能体沙箱、工具、代码和数据管道。
**English / Original**
And memory [music] coherent NVLink chip-to-chip connects GPUs directly to the fabric. Beyond GPUs, [music] NVLink chip-to-chip can scale Vera up to multiple sockets, enabling massive bandwidth between CPUs. >> [music] >> Vera delivers 1.8 times the agentic sandbox performance of X86 CPUs. Standalone Vera racks run agent sandboxes, tools, code, and data pipelines.
## 01:02:34
**中文**
Vera 与 Rubin GPU 紧密结合，保持加速的[音乐]工作流程的运行。 NVIDIA Vera BlueField 4 STS 为上下文内存和 AI 存储提供支持。 >> [音乐] >> 计算、网络、存储。 Vera 是智能体时代的 CPU。 >> [掌声] >> 这将成为我们新的主要增长动力。评论已经出来了，还不错。
**English / Original**
Tightly coupled to Rubin GPUs, Vera keeps accelerated [music] workflows moving. Nvidia Vera BlueField 4 STS powers context memory and AI storage. >> [music] >> Compute, networking, storage. Vera is the CPU for the age of agents. >> [applause] >> This is going to be our new major growth driver. The reviews are already coming out and it's pretty good.
## 01:03:10
**中文**
这是相当不错的东西。 >> [掌声] >> 现在，请记住，Grace 和 Vera 也是 AI 领域中最合格的 CPU，因为与 NVIDIA 合作进行 AI 的每个数据中心、每个云、每个企业、每个公司都已经通过了 Grace 的认证。整个软件堆栈已经针对 Grace 进行了优化。每家公司都将获得Vera资格。
**English / Original**
That's pretty good stuff. >> [applause] >> Now, remember, Grace and Vera are also the most highly qualified CPUs in the world of AI because every single data center, every single cloud, every single enterprise, every company that works with NVIDIA on AI has already qualified Grace. The entire software stack has already been optimized for Grace. Every company will be qualifying Vera.
## 01:03:49
**中文**
Vera 将成为世界上最优化的智能体 CPU。很简单，因为它将与Vera Rubin 搭配。仅仅是因为我们做出了重大的硬转变。事实上，在 Grace Blackwell 过渡期间，最大的风险是从外部 CPU x86 过渡到 Grace Blackwell。这种转变是极其危险的。但我们以令人难以置信的执行力做到了这一点。
**English / Original**
Vera will be the most optimized agentic CPU in the world. Simply because it's going to go with Vera Rubin. Simply because we made the the big hard switch. In fact, during Grace Blackwell transition, the biggest risk was going from external CPU x86 into Grace Blackwell. That transition was extremely dangerous. But we did it with incredible execution.
## 01:04:16
**中文**
现在，格蕾丝实际上是格蕾丝布莱克威尔的代名词。当人们说布莱克威尔时，他们说的是Grace Blackwell。因为它现在已经无处不在了。每个公司的软件堆栈都已针对它进行了优化。每个人的安全堆栈都已针对它进行了优化。现在Vera来了。我对此感到非常兴奋。现在看一些性能数据。
**English / Original**
Now Grace is literally synonymous with Grace Blackwell. When people say Blackwell, they say Grace Blackwell. Because it is utterly now everywhere. Every company's software stack has been optimized for it. Everybody's security stack has been optimized for it. And now here comes Vera. I'm super excited about that. Now look at some of the performance numbers.
## 01:04:39
**中文**
加速是一回事。加速 SQL 是极其困难的。 SQL 是有史以来最著名的领域特定语言 DSL。在 SQL 之前，在 CUDA 之前，就有 SQL。在 OpenGL 之前，有 SQL。由IBM发明。如今，它已成为地球上的结构化数据库引擎。每个人都使用 SQL。这是 SQL 运行速度的三倍。
**English / Original**
Speed-ups is one thing. It is extremely hard to speed up SQL. SQL the most famous domain-specific language DSL that has ever been created. Before SQL you know, before CUDA, there was SQL. Before OpenGL, there was SQL. Invented by IBM. Today, it is the structured database engine of the planet. Everybody uses SQL. This is SQL running three times faster.
## 01:05:14
**中文**
不是快 10%，也不是快 25%，而是数倍乃至 10 倍的提升，令人难以置信。下一项是实时流处理。要记住，你的 AI 不只是阅读文档，它还会持续监测遥测数据，尤其是在工厂或证券交易所内。
**English / Original**
Not 10% faster, not 25% faster, 10 times faster three times faster. Incredible. This is real-time the next one is real-time stream processing. Remember, your AI is going to be not just reading documents. Your AI is going to be watching for telemetry, especially inside a factory, inside a stock exchange. You're going to be looking for telemetry continuously.
## 01:05:44
**中文**
传入的数据突发进入 CPU。这是为纽约证券交易所运行实时流处理的 Vera CPU。纽约证券交易所总裁林恩·马丁 (Lynn Martin) 非常慷慨地与我们合作。该系统在全球范围内实时运行实时流处理。 Vera CPU 的六倍都是因为单个单线程指令执行的带宽、内核之间的内部带宽、外部的带宽。
**English / Original**
The burst of data that's coming in goes into a CPU. This is Vera CPU running real-time stream processing for New York Stock Exchange. Lynn Martin, the president of New York Stock Exchange, has been so gracious to partner with us. This system is run all over the world in real-time real-time stream processing. Vera CPU six times all because of the bandwidth the single single-threaded instruction execution, the bandwidth inside between the cores, the bandwidth outside.
## 01:06:17
**中文**
Vera是完全革命性的。那是Vera。 >> [掌声] >> 你知道，当你谈论 GPU 时，你会谈论 X 因素。很少有人谈论与 CPU 相关的实际工作负载的 X 因素。所以，我为这个团队感到非常自豪。你们做得非常好。我们即将制定一份非凡的路线图。
**English / Original**
Vera is completely revolutionary. That's Vera. >> [applause] >> You know, X factors is something you you talk about when you're talking about GPUs. It is quite rare that somebody talks about X factors on real workload real workload that is associated with CPUs. So, I'm so proud of the team. You guys did such a great job. We have an extraordinary road map coming.
## 01:06:48
**中文**
但是，真正令人兴奋的是几乎每个人都支持Vera。他们和我们一样兴奋。这是Vera的开放，它开辟了一个全新的市场。智能体智能体是一个新的工作量。过去我们为人类制造 CPU。我们需要用于智能体、智能体系统的 CPU。他们的属性不同。为什么旧的CPU会是一样的？
**English / Original**
But, what's really exciting is almost everybody is supporting Vera. They're as excited as we are. This is Vera opening up it's opened up a brand new market. Agents agents is a new workload. We built CPUs for humans in the past. We need CPUs for agents, agentic systems. Their properties are different. Why would the old CPUs be the same?
## 01:07:17
**中文**
我们将制造数以百万计的 Vera，数百万台 Vera。与我们共同把它推向市场的，包括台湾的 ODM、计算机制造商以及所有 OEM；这里还可以看到首批采用者，他们都是智能体公司。这是一个过去从未存在的新市场的开端。它不会挤占旧市场，而是开辟一个全新市场。
**English / Original**
We are building millions and millions of errors. Millions of errors. And to go to market with us, Taiwan's ODMs and computer makers, all the OEMs, and you could see the early adopters. The early adopters are the agentic companies. This is the beginning of a new market, a market that never existed before. It's not going to take away from the old markets, but this is a new market.
## 01:07:49
**中文**
面向智能体的 CPU。这个市场一定会比上一个市场更大，因为智能体的数量将远远超过人类，而且智能体非常“没有耐心”。这就是 NVIDIA Vera CPU，谢谢。>> [掌声] [掌声] >> 这其实是最重要的一页，也是核心结论。
**English / Original**
CPU for agents. And this will this this market will surely be larger than the last. And the reason for that is because there'll be a lot more agents than there are people. And then there'll the agents are very impatient. So, Nvidia Vera CPU. Thank you. >> [applause] [applause] >> This is the most important slide, really. This is the takeaway.
## 01:08:17
**中文**
这里的要点是，这就是应用程序模式。这是未来十年的计算模式。智能体、利用、编排大型语言模型。每个公司都会运行它。每家公司都将成为智能体公司。每个公司都会有智能体在里面运行。每家公司都会发现智能体需要自己的操作系统。
**English / Original**
The takeaway here is that this is the application pattern. This is the computing pattern of the next decade. Agents, harnesses, orchestrating large language models. Every company will run it. Every company will be an agent company. Every company will have agents running inside. Every company will see that agents will need its own operating system.
## 01:08:49
**中文**
每家公司都在问我们：“我们如何安全地运行智能体？我们如何为我们自己的工作负载构建智能体？”因此，我们拥有用于企业人工智能的 NVIDIA 智能体工具包。你已经看到我在众目睽睽之下建造了这个。如您所知，几乎 NVIDIA 在每次 GTC 上所做的一切，如果您回头看看我 5 年前或 10 年前的 GTC，您今天就会看到。
**English / Original**
Every company's asking us, "How do we run agents safely? How do we build agents for our own workloads?" And so, we have the Nvidia agent toolkit for enterprise AI. You've seen me build this in plain sight. Almost everything that NVIDIA does, as you know, at every GTC, if you go back and look at my GTC 5 years ago or 10 years ago, you will see today.
## 01:09:16
**中文**
你已经看到我谈论这个问题好几年了，因为我们一直在为这一刻而努力。为了构建智能体即服务或构建智能体进行运营，公司需要满足四件事。您需要的第一件事是需要模型。当然，大语言模型，越智能越好，越便宜越好，越快越好。
**English / Original**
This you've seen me talking about for several years now because we've been building for this moment. There are four things that companies need in order to build agents as a service or build agents to operate. The first thing you need is you need models. Of course, large language models, the smarter the better, the cheaper the better, the faster the better.
## 01:09:42
**中文**
第二，你需要一个编排框架来协调全局。第三，模型需要使用工具，而工具附带相应的技能。我刚才展示了 CUDA-X 库，它们将成为未来智能体极其强大的工具。最后，你需要运行时，也就是把所有部分组合起来的操作系统。这就是 NVIDIA 面向智能体的工具包。
**English / Original**
The second is you need a harness to orchestrate the whole thing. The third, these eight these models want to use tools. And these tools come with it skills. And I showed you CUDA X libraries, those are going to be amazing tools for the agents in the future. And then lastly, you need a runtime. You need the operating system that holds it all together. This is the NVIDIA toolkit for agents.
## 01:10:12
**中文**
它包括您可以修改的模型。 NVIDIA世界级的开放模型，我想向大家展示更多。您可以运行任何人的智能体。你可以运行Claude Code，不可思议的智能体，CodeX，不可思议的智能体。您可以在名为 Open Shell 的工具中运行它，这对于您的企业内部来说是高度安全的。
**English / Original**
It includes it includes models that you can modify. NVIDIA's world-class open models, I want to show you more. You can run agents from anybody. You could run uh Claude code, incredible agent, CodeX, incredible agent. You could run it inside this harness called Open Shell, which will be highly secure for your inside the enterprise.
## 01:10:38
**中文**
shell 保护智能体，使其基于安全策略。隐私受到保护。它的权利和特权是被赋予的。其身份受到保护。因此，这种 Open Shell 正在世界各地被采用。 NVIDIA OpenShell 是开源的。你会看到很多公司采用它。 Red Hat、Canonical、Microsoft，它将被到处采用。
**English / Original**
The shell protects the agent, keeps it grounded in security policies. Privacy is protected. Its rights and privileges are given. Its identity is protected. And so, this Open Shell is being adopted all over the world. NVIDIA Open Show is open source. You're going to see so many companies adopt it. Red Hat, Canonical, Microsoft, it's going to be adopted everywhere.
## 01:11:07
**中文**
这是一个重要的，这是运行时。并且该运行时针对 NVIDIA AI 平台进行了全面优化。这是无处不在的。因此，您可以在任何云、本地甚至设备上运行 OpenShell。现在，您已经拥有了他们可以使用的工具和库。您拥有可以修改或按原样使用的模型。或者你有智能体。
**English / Original**
This is an important This is the runtime. And this runtime is fully optimized for the NVIDIA AI platform. Which is everywhere. So, you can run Open Show in any cloud, on prem, and even on device. So, you have you have now tools and libraries that they can use. You have models that you can modify or use as is. Or you have agents.
## 01:11:35
**中文**
这是 OpenClaw，Hermes，另一个令人难以置信的编排框架。这些智能体工具现在可以在本地或任何地方为您运行。好的？因此，有四件事代表了现代企业的操作系统。现在，我们如何使用它？我最喜欢的智能体用例之一是芯片设计师。这是 NVIDIA 所做的最重要的事情。
**English / Original**
This be Open Claw, Hermes, another incredible another incredible harness. These agent agentic harnesses can now run on prem or for you anywhere. Okay? So, four things and this represents the operating system of the modern enterprise. Now, how do we use this? One of my favorite use cases of agents is chip designers. It is the single most important thing that NVIDIA does.
## 01:12:07
**中文**
所以，当然，我们必须与 Cadence 合作来构建超级智能体。芯片设计超级智能体。它由 Codex 或 Claude Code 编排。它有 RTL 和架构图或原理图或呃规范作为输入以及您需要修复的任何内容。我们一起创建了一些超级智能体，这些超级智能体针对 Nemotron 的 NVIDIA 运行时进行了优化。
**English / Original**
And so, of course, we have to partner with Cadence to build super agent. A chip design super agent. It is orchestrated by Codex or Cloud Code. It has RTL and architecture diagrams or schematics or uh specifications as input and whatever you need to fix. And together, we created some super agents that are optimized for the NVIDIA runtime with NeMoTron.
## 01:12:41
**中文**
来看一下，确实令人惊叹。>> Cadence 与 NVIDIA [音乐] 正在携手打造芯片设计智能体。数十万颗 NVIDIA 芯片组合在一起，构成驱动世界前沿 AI 模型的 AI 工厂。设计这些芯片及其运行的系统，[音乐] 是工程领域最艰巨的挑战之一。
**English / Original**
And let's take a look. It's really incredible. >> Cadence and NVIDIA [music] are partnering to build chip design agents. Hundreds of thousands [music] of NVIDIA chips come together to make the AI factories that power the world's frontier AI models. Designing these chips and the systems [music] they run in is one of the hardest engineering challenges.
## 01:13:05
**中文**
数万亿个晶体管、三维电路、微观[音乐]规模。与皮秒同步的每个门、每根电线都必须完美协调地工作，不允许有任何误差。物理原型太慢而且成本太高。因此，工程师[音乐]在数字领域工作。每个芯片最初都是一组架构规范，然后转化为 RTL（芯片设计语言）。
**English / Original**
Trillions of transistors, three-dimensional circuits, microscopic [music] scale. Every gate, every wire synchronized to picoseconds must work in perfect harmony with no margin for error. Physical prototypes are too slow and too costly. So, engineers [music] work in the digital realm. Each chip begins as a set of architectural specifications, then translated into RTL, the language of chip design.
## 01:13:32
**中文**
RTL 必须在仿真中进行验证。一个错误可能会使芯片延迟数月。 >> [音乐] >> 在 NVIDIA，数千名工程师，每年数十亿计算机小时，数百万次测试 >> [音乐] >> 编写、运行和调试。这个周期需要团队几周的时间。为了压缩这个[音乐]周期，Cadence 和 NVIDIA 构建了一个设计验证智能体。
**English / Original**
RTL must be verified in simulation. A single bug can delay a chip by months. >> [music] >> At NVIDIA, thousands of engineers, billions of computer hours per year, millions of tests >> [music] >> written, run, and debugged. A cycle that takes teams weeks. To compress this [music] cycle, Cadence and NVIDIA built a design verification agent.
## 01:13:53
**中文**
Codex 负责协调这一流程。Cadence ChipStack 启动 RTL 验证循环，[音乐] 由 Nemotron 驱动，并由 NVIDIA OpenShell 提供安全保护。它会调用专门负责 RTL 生成、测试平台创建、回归测试和调试的子智能体。[音乐] 系统自主推进；ChipStack 智能体使用 Cadence Xcelium 运行数百次仿真，并使用 Jasper 开展形式验证。[音乐]
**English / Original**
Codex orchestrates the process. Cadence ChipStack launches the RTL verification [music] loop. Powered by Nematron and secured by NVIDIA Open Shell. Calling on expert [music] sub-agents in RTL generation, test bench creation, regression testing, and debug. The system drives itself. [music] The ChipStack agents run hundreds of simulations with Cadence Xcelium, formal verification [music] with Jasper.
## 01:14:19
**中文**
设计缺陷暴露无遗。代码[音乐]中的错误已修复。以前需要几周的时间现在只需几个小时。验证周期快了 40 倍以上。 NVIDIA 和 Cadence [音乐] 正在携手利用 AI 智能体重塑芯片设计。 >> 从几周到几个小时。 >> [掌声] >> 从几周到几小时。从几周到几小时。 NVIDIA 拥有数千名芯片设计师。我们将聘请数十万Cadence超级智能体与我们合作，以便我们能够加速我们的公司，使我们能够更加雄心勃勃，创造更加惊人的事物，跑得更快。你看到了
**English / Original**
Design flaws revealed. Bugs in the code [music] fixed. What once took weeks now takes hours. Verification cycles over 40 times faster. Together, NVIDIA and Cadence [music] are reinventing chip design with AI agents. >> From weeks from weeks to hours. >> [applause] >> From weeks to hours. From weeks to hours. NVIDIA has thousands of chip designers. We are going to hire hundreds of thousands of Cadence super agents that work with us so that we can accelerate our company so that we can be even more ambitious, create even more amazing things, run even faster. You saw
## 01:15:03
**中文**
前面展示的工具包包含模型、编排框架和工具。本例中的工具是 Cadence 模拟器、验证器以及形式验证系统。智能体没有耐心，要求立即得到答案，因此我们与 Cadence 密切合作，用 CUDA 加速其全部工具。于是整套系统包括模型、编排框架、CUDA 加速的库与工具，以及运行时。
**English / Original**
earlier that the toolkit with models harness tools The tools in this case are Cadence simulators and verifiers, formal verification systems. It is the reason why we're working with Cadence so hard to accelerate all of their tools on CUDA because the agents are impatient. The agents want the answer immediately. And so models harnesses accelerated CUDA accelerated libraries and tools and then the runtime.
## 01:15:37
**中文**
你刚才看到的就是所有这些结合在一起的。现在，它开始的一件事是一个伟大的模型，Cadence 可以对其进行修改和调整，使其成为 Cadence 工作流程、Cadence 专业知识的专家，以便他们可以利用其专有知识创建 Cadence 专有的超级智能体。他们必须从一个优秀的模型开始。我们称之为 Nemotron。
**English / Original**
What you saw just now is all of that coming together. Now, one of the things that it starts with is a great model that Cadence could modify and tune to be expert at the Cadence workflow, at the Cadence expertise so that they could create super agents that are proprietary to Cadence with their proprietary knowledge. They have to start with an excellent model. We call it Neumotron.
## 01:16:06
**中文**
NVIDIA 致力于为世界构建开放模型，以便你们所有人、我们所有人都可以创建我们自己的智能体。今天我们发布了 Nemotron 3 Ultra。是的，我们的下一个开放模型。而且它很聪明。 >> [掌声] >> Nemotron 模型不仅为您提供模型，我们还为您提供用于训练模型的所有数据。因为我们拥有一个由令人难以置信的合作伙伴组成的联盟，所以您可以在这里看到我们所有的合作伙伴。
**English / Original**
Nvidia is dedicated to build open models for the world so that all of you, all of us could create our own agents. Today we're announcing the Neumotron 3 Ultra. Yep, our next open model. And it is smart. >> [applause] >> The Neumotron models not only give you the model we give you all the data that we used to train the model. And because we have a coalition of incredible partners, you can see all of our partners down here.
## 01:16:46
**中文**
我们一起工作，互相贡献数据。 Nemotron 经过世界上最大的长期运行推理模型、长期运行工具任务解决、使用数据集的工具套件之一的训练。由于我们所有良好的合作伙伴关系，所有这些来自模型、训练脚本和数据的内容都完全可供您使用。这是最好的开放模型。
**English / Original**
We work together, contribute data to each other. NeMoTron is trained on one of the largest suites of long-running reasoning models, long-running tool task solving, tool using data sets in the world. Because of all of our great partnerships, all of this from the model, the training script, and the data made all completely available to you. This is open models at its best.
## 01:17:16
**中文**
这是世界上最好的开放模型体系。目标很简单：让你把整套系统拿走，在其上增加能力、改进它，并把它变成自己的模型。Nemotron 3 Ultra 速度提升 5 倍；它是全球首个采用混合 SSM（状态空间模型）架构的混合专家模型。这个架构本来就很快，我们又让它更快，使你能够快速思考；思考越快，在相同成本下就能思考得越久。
**English / Original**
The best open model system policies in the world. Simple goal is so that you can take all of it, add to it, make it even better, make it yours. NeMoTron 3 Ultra is five times faster. This is the world's first model based on a hybrid architecture of SSM, state space models, with mixture of experts. The architecture is incredibly fast. We made it faster that you could think fast. When you think fast, you could think longer at the same cost.
## 01:17:51
**中文**
因此，它的速度提升了 5 倍；即使与全球最具成本效益的模型相比，按总 FLOPs 和总推理时间计算，运行成本也降低了 30%。我们的比较对象是全球最好的开放模型：前沿级智能，快 5 倍，成本低 30%，而且完全开放。我们对此全力投入。现在是 Nemotron 3，我们已经在开发 Nemotron 4。
**English / Original**
So, five times faster. It is also 30% cheaper, 30% lower cost to run in total flops, in total inference time, than even the most cost-effective in the world. We're comparing against the world's best open models. Frontier smart. Five times faster. 30 point 30% cheaper. Completely open. We're completely dedicated to this. This is now NeMoTron 3. We're currently working on NeMoTron 4.
## 01:18:25
**中文**
因此，包括模型、编排框架、工具和技能以及运行时在内的整个工具包是世界上每个企业公司现在都有能力创建自己的智能体的原因，就像 Cadence 对其超级智能体所做的那样。我们正在与很多公司合作，包括 Cadence 和 CrowdStrike、Dassault 和 Palantir、SAP 和 ServiceNow。人们总是说，“黄仁勋，智能体会扰乱这些市场。”
**English / Original**
So, this entire toolkit from models, harnesses, tools and skills, and runtimes is the reason why every enterprise company in the world has the ability now to create their own agents just like Cadence did with their super agents. And we're working with so many companies, Cadence and CrowdStrike and Dassault and Palantir, SAP and ServiceNow. People were always said, "Jensen, the agents are going to disrupt these markets."
## 01:18:58
**中文**
我说的完全相反，你现在可以看到了。智能体将为我的合作伙伴和朋友创造有史以来最大的机会。我们有 Nemo（用于企业 AI 的 NVIDIA 智能体工具包）来帮助他们。那么，就这样吧。 >> [掌声] >> 首先，Vera Rubin 全面投入生产。二、Vera CPU，专为新一代智能体打造的CPU。第三，NVIDIA 的企业 AI 工具包使得
**English / Original**
I said completely opposite and you can now see it. Agents is going to create the largest opportunity ever for my partners and friends. And we have the Nemo the the NVIDIA agentic toolkit for enterprise AI to help them. So, there you go. >> [applause] >> First, Vera Rubin in full production. Two, Vera CPU, CPU built for a new generation for agents. And three, NVIDIA's enterprise AI toolkits so that
## 01:19:38
**中文**
每个企业、每个企业软件公司都可以建立智能体。 >> [掌声] >> 我和你的关系就是从这里开始的。你们中的许多人，我在台湾的许多朋友和合作伙伴，你们的公司都是从这里开始的。从很多方面来说，这都是现代计算机行业 40 周年的开端。
**English / Original**
every enterprise and every enterprise software company can build agents. >> [applause] >> My relationship with you started here. And many of you, many of you, many of my friends and partners here in Taiwan, your companies started here. This is in a lot of ways the beginning of the modern computer industry, 40 years now.
## 01:20:14
**中文**
NVIDIA已经33岁了。 PC 行业已经开始转向 Windows 1 和 Windows 2 以及 Apple Apple 1 和 Apple 2，当我们出现时，Windows 3.1 就是 PC。如您所知，Windows 95 使 PC 变得个性化。它将个人电脑从企业、公司手中夺走，并使其成为消费电子设备。
**English / Original**
NVIDIA's 33 years old. The PC industry was already starting to get to Windows 1 and Windows 2 and Apple Apple 1 and Apple 2 and by the time that we came along, Windows 3.1 was the PC. And as you know, Windows 95 made PC personal. It took PC from enterprises, com- companies, and made it into a consumer electronics device.
## 01:20:43
**中文**
每个人都应该拥有一个，而且每个人都这样做了。这就是开始。这个计算平台做了一些非常聪明的事情。如您所知，Windows 不仅仅是分解的。 Windows 已被正确抽象。它的架构恰到好处。系统 BIOS、开放芯片组、带有驱动程序的操作系统、可以在运行时连接和安装的驱动程序。
**English / Original**
Everybody should have one, and everybody does. This is the beginning. This computing platform did several things incredibly smart. Windows was not just disaggregated, as you know. Windows was properly abstracted. It was architected just right. Systems BIOSes, open chipsets, the operating system with drivers, drivers that could be connected and installed at runtime.
## 01:21:15
**中文**
还有一个具有多媒体 API 的抽象层，它使 PC 向我们今天所熟知的领域开放。这些元素中的每一项对于个人电脑如此受欢迎都是至关重要的。 40 年后，微软和NVIDIA将重塑 PC。这将是新电脑。现在，明天晚上，明天晚上，我想明天晚上就是我们的时间了，但我要和萨蒂亚在一起。我们将更多地讨论我们正在一起做的工作。
**English / Original**
And an abstraction layer with a multimedia API that was that opened up the PC to what we all know today. Each one of these elements were essential in making the PC so popular. 40 years later, Microsoft and Nvidia are going to reinvent the PC. This is going to be the new PC. Now, tomorrow night, tomorrow night, I think it's tomorrow night our time, but I'm going to be with Satya. We're going to talk a lot more about the work that we're doing together.
## 01:21:53
**中文**
微软和NVIDIA在过去 3 年里花了很长时间彻底重塑了 PC 的工作方式，以便我们为这一刻做好准备。正如我之前提到的，称为智能体的计算模式将在AI 云中运行。它将在企业内部运行。它也将在您的 PC 上运行。
**English / Original**
Microsoft and Nvidia, over the last 3 years, it took this long to completely reinvent how the PC's going to work, so that we could be ready for this moment. As I mentioned earlier, that compute pattern called the agent is going to run in AI clouds. It's going to run inside enterprises. It is also going to run on your PC.
## 01:22:18
**中文**
当该 PC 拥有自主智能体时会发生什么？一个能够帮助您、理解您的智能体。你可以和它说话。它可以看着你。你可以要求它重新归档，帮你做一些研究。我将向您展示它可以做更多事情。但新操作系统当然是旧操作系统加上大语言模型。
**English / Original**
What's going to happen to that PC when it has an autonomous agent? An agent that's helping you, that understands you. You could talk to it. It could look at you. You could ask it to refile those, go help you do some research. It could do a lot more that I'll show you. But the new operating system is of course the old operating system plus large language models.
## 01:22:45
**中文**
大型语言模型在很多方面都是 DirectX 的现代版本。它当然有输入和输出，理解提示，它理解计算机视觉，它可以生成视频，它可以生成声音。它是个人计算机的现代延伸、智能延伸。最重要的是，正如我之前提到的，该应用程序现在将在智能体运行时中被替换。
**English / Original**
Large language models in a lot of ways is the modern version of DirectX. It has of course input and output, understands prompts, it understands computer vision, it can generate video, it can generate sounds. It is the modern extension, the intelligence extension of the PC, of a computer. On top of that, the application, as I mentioned before, is going to be replaced by now in agentic runtime.
## 01:23:15
**中文**
这就是现代应用程序和智能体。现在让我们看看它能做什么。 >> 它始于一个火花。随着人工智能时代的到来，40 年来首次重新构想 PC 的想法。在智能体世界中，我们的个人[音乐]计算机会变成什么样子？智能体在本机运行，连接到本地或云中的模型。我们的个人[音乐]人工智能，沙盒化以确保安全，持续运行，完成工作。
**English / Original**
And that is the modern application and agent. Let's now take a look at what it can do. >> It started with a spark. An idea to reimagine the PC for the first time in 40 years with the age of AI. What becomes of our personal [music] computer in a world of agents? Agents running natively, connected to models, local or in the cloud. Our personal [music] AI, sandboxed for security, running continuously, getting work done.
## 01:23:52
**中文**
芯片和操作系统必须[音乐]不断发展。隆重推出 RTX Spark。我们将 33 年[音乐]年来学到的一切都浓缩到了一块芯片中。 Blackwell RTX GPU 具有 6,144 个 CUDA 核心，1 petaflop 的 AI 性能，与 MediaTek 合作构建的定制 20 核 [音乐] Grace CPU，由 NVLink 融合。 128 [音乐] GB 统一内存。台积电3纳米工艺。
**English / Original**
The chips and the OS must [music] evolve. Introducing RTX Spark. Everything we've learned over 33 [music] years distilled into one chip. Blackwell RTX GPU with 6,144 CUDA cores, 1 petaflop of AI performance, a custom 20-core [music] Grace CPU built in partnership with MediaTek, fused by NVLink. 128 [music] GB of unified memory. TSMC 3-nanometer process.
## 01:24:29
**中文**
700亿个晶体管。 [音乐] 并与微软密切合作，推出Windows平台智能体。我们正在重塑[音乐]个人电脑。为了创造。用于游戏。对于智能体来说。 [音乐] 这是新的个人计算革命的曙光。它始于 NVIDIA [音乐] RTX Spark。 >> [掌声]
**English / Original**
70 billion transistors. [music] And in close collaboration with Microsoft, a Windows platform for agents. We're reinventing [music] the personal computer. For creating. For gaming. For agents. [music] This is the dawn of a new personal computing revolution. And it starts with NVIDIA [music] RTX Spark. >> [applause]
## 01:25:14
**中文**
>> 就是这里。
**English / Original**
>> Here it is.
## 01:25:16
**中文**
当然，我要向你们展示最美丽的部分，那就是电子游戏。它也是最贴近我们内心的。这是极限竞速。顺便说一句，这是007。 007的新游戏，我很期待玩。我看起来有点像他。女士们、先生们，请介绍 NVIDIA 的 RTX Spark 笔记本电脑。现在，>>[掌声]>>谢谢。
**English / Original**
Of course, I got to show you the most beautiful part, which is video games. It is It's also the closest to our heart. This is Forza. This is 007, by the way. The new 007 game, I'm looking forward to playing it. I look a little bit like him. Ladies and gentlemen, NVIDIA's RTX Spark laptops. Now, >> [applause] >> thank you.
## 01:25:51
**中文**
我口袋里的东西太多了。好的。好的。这是世界上有史以来最令人惊奇的芯片。这是我们与联发科合作打造的 N1X。我想我早些时候见过瑞克。这是N1X。这是一个漂亮的芯片。这是一款坦率地说需要 33 年才能打造出来的芯片。原因是 100% 的 NVIDIA 软件堆栈都在这里运行。
**English / Original**
I have too many things in my pocket. Okay. All right. This is the most amazing chip the world has ever built. This is the N1X that we built in partnership with MediaTek. I think I saw I saw Rick earlier. This is N1X. This is a beautiful chip. And this is This is a a a chip that frankly would take 33 years to build. And the reason for that is because 100% of Nvidia software stack runs here.
## 01:26:25
**中文**
如果您想运行数字生物学，没问题。如果你想做地震处理，没问题。你想做天体物理学，没问题。与 CUDA 相关的一切，所有物理学、所有生物学、所有基因组学、所有人工智能，都没问题。全部电脑绘图，没问题。 NVIDIA 创建的每一个应用程序以及 Windows 运行的每一个应用程序。
**English / Original**
If you want to run digital biology, no problem. If you want to do seismic processing, no problem. You want to do astrophysics, no problem. Everything associated with CUDA, all the physics, all the biology, all the genomics, all the AI, no problem. All the computer graphics, no problem. Every single application Nvidia has ever created and every single application that Windows has ever run.
## 01:26:52
**中文**
微软和NVIDIA精心优化了一切，使这台计算机真正运行了世界上创造的一切，而且现在还运行智能体。一台令人难以置信的计算机。我为此感到非常自豪。 >> [掌声] >> 好的。现在我希望您在下一个视频中记住这一点，我将向您展示。想象一下这里的所有内容都将在您的 PC 上运行。现在该计算机可以具有本地 Nemotron 3 ultra 模型或
**English / Original**
Microsoft and Nvidia meticulously optimized everything so that this computer literally runs everything the world has ever created plus it now runs agents. An incredible computer. I'm so proud of it. >> [applause] >> Okay. Now I want you to keep that in mind in the next video I just I'm going to show you. Just imagine everything here is going to run on your PC. Now that computer could have a local Neumotron 3 ultra model or
## 01:27:32
**中文**
Nemotron 3 超级模型，或者它可能有 Claude 代码或 Codex 或云中或网络上的其他模型，它将会工作并做出一些令人惊奇的事情。让我们玩。 >> 每栋房子都是从一个想法开始的。从想法[音乐]到设计需要大量的工具、专业知识和大量的[音乐]时间。
**English / Original**
Neumotron 3 super model or it could have Claude code or Codex or some other model in the cloud or something on the network and it's going to it's going to work and do something amazing. Let's play. >> Every house starts as an idea. Getting from idea [music] to design takes a myriad of tools, expertise, and a lot [music] of time.
## 01:27:59
**中文**
现在，在 RTX Spark [音乐] 上本地运行的智能体可以帮助我使用笔记本电脑上的工具设计一座房子，该工具带有一个OpenShell沙箱，运行与云中的 Claude Sonnet 连接的 [音乐] Hermes 编排框架。我选择该网站，分享我的概念草图和风格情绪板来激发我的设计灵感。 >> [音乐] >> 以及提示，需求和设计意图的文字描述。
**English / Original**
Now an agent running locally on RTX Spark [music] can help me design a house using the tools on my laptop with an open shell sandbox running the [music] Hermes harness connected to Claude Sonnet in the cloud. I select the site, share my concept sketches and mood board of styles to inspire my design. >> [music] >> And the prompt, a text description of the requirements and the design intent.
## 01:28:28
**中文**
我的智能体开始工作。它使用笔记本电脑上的工具打开 Rhino，对场地建模，塑造地形、退界范围和建筑围护结构；随后提出针对成本、舒适度与品质优化的建筑形态。[音乐] 形态确定后，智能体生成室内布局，墙体、动线与房间逐渐成形。
**English / Original**
My agent goes to work. Using [music] the tools on my laptop, it opens Rhino and starts modeling the site, shaping terrain, setbacks, and the building envelope. Then it proposes building [music] forms optimized for cost, comfort, and quality. With the form defined, my agent generates the interior layout. Walls, circulation, rooms begin to take shape.
## 01:28:53
**中文**
每当我想要[音乐]调整、改变时我就会加入。门、>> [音乐] >> 窗户和结构元素会自动放置。我的智能体检测到自己的[音乐]错误并修复它们。当我批准后，智能体将模型从 Rhino [音乐] 导出到 Blender 中。材料和对象属性随设计上下文完好无损地转移。
**English / Original**
I jump in whenever I want to [music] adjust, to change. Doors, >> [music] >> windows, and structural elements are placed automatically. My agent detects its own [music] mistakes and fixes them. When I approve, the agent exports the model from Rhino [music] into Blender. Materials and object properties transfer with the design context intact.
## 01:29:21
**中文**
我微调[音乐]素材，使外观恰到好处。然后我挑选镜头。 Blender 渲染房子。我的智能体使用生成式 AI 和 Flux 2 模型，使它们变得像照片一样真实。多个视点、照明条件。曾经复杂的[音乐]工作流程现在由我的智能体在 RTX Spark 上进行指导和简化。
**English / Original**
I fine-tune the [music] materials, get the look just right. Then I pick the shots. Blender renders the house. My agent, using generative AI with the Flux 2 model, makes them photo real. Multiple viewpoints, lighting conditions. What was once a complex [music] workflow is now guided and simplified by my agent, working with on RTX Spark.
## 01:29:45
**中文**
以想象的速度[音乐]进行设计。 >>[掌声]>>PC世界的智能体。开发商对此非常兴奋。这是一台令人难以置信的计算机。所有的加速，所有与之相关的软件功能，与每个开发人员合作，让所有人都难以置信。下一个，Adobe。
**English / Original**
Design at the speed [music] of imagination. >> [applause] >> PC in the world of agents. The developers are so excited by this. This is an incredible computer. All of the acceleration, all the software capabilities associated with it, working with every developer to make it incredible for all of you. The next one, Adobe.
## 01:30:14
**中文**
当然，这是令人难以置信的工具套件，全世界有数以千万计的人在使用。他们重新设计了 Adobe Photoshop 和 Premiere 的核心架构，并将其发布到 RTX Spark。它的速度是原来的两倍。已经很快了。现在，速度将提高一倍。而且它还被设计为对智能体友好。借助其 MCP 服务器，它现在可以与笔记本电脑上的智能体进行交互。
**English / Original**
Incredible tool suite, of course, used by tens of millions of people around the world. They have re-engineered the architecture, the core of Adobe Photoshop and Premiere, and they're releasing it for RTX Spark. It is twice as fast. It's already fast. Now, it's going to be twice as fast. And it It's also designed to be agent-friendly. With its MCP server, it can now interact with agents on your laptop.
## 01:30:40
**中文**
如此兴奋地将 RTX RTX Spark 推向市场的客户数量和合作伙伴数量令人难以置信。要知道，这是 40 年来 PC 重塑系列中的第一次。我很高兴你们所有人以及世界各地的生态系统都加入了我们。这基本上是所有人。每个人都将使用 RTX Spark，并与我们所有人一起打造极其智能、强大且美观的笔记本电脑。非常感谢。
**English / Original**
The number of customers, the number of partners that are so excited to bring RTX RTX Spark to the market is just incredible. You know, this is the first across the lineup of PC reinvention for 40 years. And I'm just so happy that all of you and the ecosystem around the world has joined us. This is basically everybody. Everybody will RTX Spark and will be building incredibly smart and powerful and beautiful laptops with all of us. Thank you very much.
## 01:31:15
**中文**
>> [掌声] >> 但这还不是全部。这还不是全部。 RTX Spark 是对笔记本电脑的重新发明。但事实上，微软和NVIDIA正在重塑整个 PC。今天，我们宣布推出全新系列。三款革命性的 Windows 机器，涵盖台式机、笔记本电脑和工作站。全部 100% Windows 兼容、100% CUDA、100% NVIDIA AI Tensor Core。
**English / Original**
>> [applause] >> But that's not all. That's not all. RTX Spark is a reinvention of laptop. But in fact, Microsoft and Nvidia is reinventing all of PC. And today, we're announcing a whole new line. Three revolutionary Windows machines covering desktop, laptop, and workstations. All 100% Windows compatible, 100% CUDA, 100% Nvidia AI Tensor Core.
## 01:31:52
**中文**
您所看到的在全球所有不同平台上的 NVIDIA 上运行的所有内容都在这里运行。这是 40 年来第一条完全重新设计、重新发明的 PC 系列。现在，真正令人惊奇的是这个。这是 RTX Spark 笔记本电脑。这是桌面。所以，这个来自 MSI。
**English / Original**
Everything that runs that you see that runs on Nvidia in all these different platforms around the world runs here. This is the first completely re-engineered, reinvented line of PCs that has happened in 40 years. Now, what's really amazing is this. So, this is This is the RTX Spark laptop. This is the desktop. So, this one's from MSI.
## 01:32:19
**中文**
Joseph，这台归你了。好，看看它多漂亮。这台智能体计算机可以全天候运行，不按用量计费。你可以把自己的智能体下载下来，在这里运行你的 OpenClaw；这就是属于你的“Claw”。它始终在线，不必担心计量费用，并且可以连接整栋房子：笔记本电脑、显示器、所有摄像头、烘干机、饮水机、热水器、安全系统——任何你想接入的设备都可以连接到它。
**English / Original**
Joseph, this one's yours. Okay. Look how beautiful it is. This agent could run 24/7. Meter-free. And you could download your agent. You could raise your lobster in here. This is your claw. It's running all the time. No meter anxiety. And it's sitting here connected to your whole house. Connected to your laptop. Connected to your display. All the cameras. Your your dryer. Your water cooler. Your water heater. Your everything. Whatever you want. Your security system. All connected to this.
## 01:32:59
**中文**
这将成为你的个人人工智能。您的私人AI 智能体。随着时间的推移，它会变得越来越聪明，因为今天我们有了 Nemotron 3 Ultra。明天我们将推出 Nemotron 4，然后是 Nemotron 5，Nemotron 6。我们会不断让它变得越来越智能。与此同时，这是坐在家里帮助你做事。如果您想预订旅行，没问题。
**English / Original**
And this becomes your personal AI. Your personal AI agent. And it gets smarter and smarter and smarter over time because today we have Numatron 3 Ultra. Tomorrow we have Numatron 4. And then Numatron 5. Numatron 6. And we just keep getting it smarter and smarter and smarter. And meanwhile, this is sitting at home helping you do things. If you want to book a travel, no problem.
## 01:33:22
**中文**
如果您想要一个令人难以置信的系统，那么这是适用于 Windows 的 DGX Station。与Windows兼容。运行 Windows 中的所有内容。而且它有 768 GB 内存。因此，您可以运行一万亿个参数模型。这太令人难以置信了。 20 petflops。每秒 8 TB 的内存带宽。这个就在你的桌子旁边。基本上，如果您是大型语言模型的开发人员，那么您就是智能体的开发人员，让这个坐在旁边
**English / Original**
And if you if you want an incredible system, this is a DGX Station for Windows. Compatible with Windows. Runs everything in Windows. And And it has 768 GB of memory. And so, you could run a trillion parameter model. This is unbelievable. 20 petaflops. 8 TB per second of memory bandwidth. And this sits by your desk. You basically, if you're a developer of large language models, you're a developer of agents, having this sit by
## 01:34:04
**中文**
您的办公桌为您提供所需的所有计算，然后在部署时将其放入云中。现在，如果你看看这个并思考一下，就会发现这里正在发生一些事情。请记住，15、20 年前，我们曾经有过一个叫做电话的想法。今天，我们有一个想法，叫做个人电脑。如今，当您想到手机时，您不会用它做的一件事就是打电话。
**English / Original**
your desk gives you all the compute you need, and then when you deploy it, you put it into the cloud. Now, there's something that if you look at this and think about this, something is happening here. Remember, 15, 20 years ago, we used to have an idea called a phone. Today, we have an idea called a PC. Today, when you think about your phone, the one thing you don't do with it is make phone calls.
## 01:34:39
**中文**
你几乎做其他所有事情。因此，这款手机对您来说与过去的手机有着非常不同的意义。我确信 10 年后将会发生的事情是，十年后的 PC 和您今天想到的 PC，无论您启动应用程序、单击还是打字，都会成为一种工具，而这台 PC 将完全不同。
**English / Original**
You do just about everything else. And so, that phone means something very different to you than a phone of the past. I am certain what's going to happen here is that the PC 10 years from now and the PC that you think about today, a tool whether you launch applications, click and type, and this PC is going to be completely different.
## 01:35:06
**中文**
这是我的理论。我完全可以想象，就像现在每家每户都有家庭影院，或者很多房子都有家庭影院、大电视、割草机、洗碗机一样，我完全可以想象有一天，你家里居然有一台人工智能超级计算机。它运行着你所有的智能体，它运行着你所有的助手，他们一直在为你做各种各样的事情。
**English / Original**
Here's my theory. I can totally imagine, just as every house today has a home theater, or many houses have home theaters, big TVs, lawnmowers, dishwashers, I could totally imagine that someday, there's actually an AI supercomputer in your house. And it's running all of your agents, it's running all of your assistants, and they're doing all kinds of things for you all the time.
## 01:35:36
**中文**
而且你家里必须有它，就像你家里有家庭影院、你家里有音响、你家里有游戏机一样。您将拥有在您家中运行的辅助AI 智能体计算机。随着时间的推移，这些对你来说变得更像 R2-D2。对您而言，它更像是 C-3PO，而不是 PC。
**English / Original**
And you have to have it in your house, just like you have a home theater in your house, you have stereos in your house, you have game consoles in your house. You would have assist AI agent computers running in your house. And these in time becomes a lot more like R2-D2 to you. It becomes more like C-3PO to you than it feels like a PC to you.
## 01:36:02
**中文**
毫无疑问，计算机的重新发明与我们现在所知的智能手机的重新发明一样重要。因此，这就是旅程的开始。这是新线路的开始。因此，我们为此制定了路线图。这对我们来说是一个全新的产品系列。每一代架构都会有台式机、笔记本电脑、工作站，然后是台式机、笔记本电脑和工作站。我非常高兴、非常荣幸的是，全球 100% 的 PC 行业都加入了我们
**English / Original**
There is no question this reinvention of the computer is as big of a deal as the reinvention of the phone into what we now know as the smartphone. And so, this is the beginning of that journey. This is the beginning of a new line. And so, we have a roadmap for this. This is a brand new product family for us. Every single generation of architecture we will have a desktop, a laptop, a workstation, and then a desktop, a laptop, and workstation. And the thing that I'm just incredibly pleased, incredibly honored is that 100% of the world's PC industry has joined us
## 01:36:42
**中文**
共同重塑 PC：一条新产品线，一个新起点。谢谢。>> [掌声] >> 大家知道，智能体 AI 本质上就是数字机器人。它能够理解、推理、规划、行动并使用工具。智能体 AI 将运行在所有这些计算机上。多年来，你们已经听我逐一谈过这些系统。
**English / Original**
to reinvent the PC. A new line, a new beginning. Thank you. >> [applause] >> As you know, a gentic AI is just a digital robot. It understands, it reasons, it plans, and it acts and uses tools. A gentic AI is going to run across all of these computers. And you've seen me talk about each and every one of these over time.
## 01:37:24
**中文**
我们正在研发人形机器人计算机以及各种机器人计算机，也在研发自动驾驶汽车计算机和卫星计算机。GeForce 拥有 Tensor Core；我刚才又介绍了全新的 PC 产品线。农业设备、制造设备和重工业设备都将实现智能体化。
**English / Original**
We're working on humanoid robotics computers, robotics computers of all kinds. We're working on self-driving car computers. We're working on satellites. You have GeForce, which is has tensor cores. I just talked about a whole new line of PCs. Agriculture equipment, manufacturing equipment, heavy industry equipment will all be a gentic.
## 01:37:45
**中文**
你甚至还会有一个自己的小助手。即使是您的基站，未来的广播电台也将是智能体的。了解流量并思考如何与其他基站协调，以便可以使用尽可能少的能源，提高利用率、频谱效率的效率。
**English / Original**
You'll even have a little agentic helper for yourself. Even your base stations, the radio stations of the future are going to be agentic. Understanding traffic and thinking about how to coordinate with the other base stations so that you could use as little energy as possible, increase the utilization, the efficiency of the spectral efficiency.
## 01:38:10
**中文**
所以，一切都会运行智能体。如今，NVIDIA基本上处于中间位置。但我非常确定，随着时间的推移，将会有数百亿、数千亿的智能体系统、智能体计算机在世界各地运行。最大的问题是数据。就语言模型而言，我们训练的所有英语和互联网上的所有语言都是从我们的角度出发的。我们写了它，我们正在读它。
**English / Original**
And so, everything will run agents. Today, Nvidia is largely in the center. But I am pretty certain that there will be tens of billions, hundreds of billions over time of agentic systems, agentic computers that are going to be running around the world. The biggest problem is data. In the case of language models, all the English and all the language that we have on the internet that we trained on was from the perspective of us. We wrote it and we're reading it.
## 01:38:42
**中文**
然而，为了为人工智能、机器人技术创建数据，它必须处于机器人的感知、视角中。世界上大多数视频数据都来自第三人称，而不是第一人称。因此，智能体系统、机器人系统、物理人工智能、数据是最难的问题。你已经看到我们沿着这个阶梯向上移动。我们从远程操作开始，这基本上是人类演示。这与强化学习人类反馈的重大突破没有什么不同。
**English / Original**
However, in order to create a data for AI, robotics, it has to be in the perception, the perspective of the robot. And most of the world's video data is from a third person, not first person. And so, agentic systems, robotic systems, physical AI, the data is the hardest problem. You've seen us move up this ladder. We started with teleoperations, which is basically human demonstration. This is no different than the big breakthrough of reinforcement learning human feedback.
## 01:39:18
**中文**
然后我们使用模拟。这就是 Omniverse 发挥作用的地方。这与强化学习可验证的奖励没有什么不同。好的？因此，我们使用这些系统来引导人工智能模型，即物理人工智能模型。最终，我们能够向第三人称学习，将其重新投影到第一人称，现在最终通过引导，我们有了一个世界基础模型，可以从任何你想要的角度理解物理世界。第三人称，第一人称，从外到内，从内到外，都没关系。
**English / Original**
This then we use simulation. This is where Omniverse comes in. This is no different than reinforcement learning ver- verifiable rewards. Okay? And so, we use these systems to bootstrap the AI model, the physical AI model. Eventually, we're able to learn from third per- third person, reprojecting it into first and now eventually through bootstrapping we have a world foundation model that can understand the physical world from any perspective you want. Third third-person, first-person, outside in, inside out, doesn't matter.
## 01:39:58
**中文**
这确实是一个重大突破，今天我们发布了 Cosmos 3。Cosmos 3 是物理人工智能的前沿。我们处于语言模型的前沿。有很多人在努力。然而，在物理人工智能方面，我们绝对是世界上最好的。我为团队做到这一点感到非常自豪。这是您所有工作的基础模型。
**English / Original**
This is a big breakthrough indeed and today we are announcing Cosmos 3. Cosmos 3 is the frontier of physical AI. We are at the frontier with language models. There's so many people working on it. However, in physical AI, we are absolutely the world's best. I am so proud of the team for doing this. This is the foundation model for all of your work.
## 01:40:28
**中文**
每当你想创造一个机器人，每当你想创造一个工厂机器人或在工厂工作的机器人，任何涉及物理世界的机器人，你现在都有一个同伴，一个可以理解和推理的 Cosmos 3。就可以生成.它可以在循环中进行模拟。它甚至可以是策略本身。它在世界各地的排行榜上名列前茅。我为 Cosmos 感到无比自豪，今天我们宣布推出 Cosmos 3。让我们来看看。
**English / Original**
Whenever you want to create a robot, whenever you want to create a factory robot or a robot that works in a factory, any kind of robot there in that involves physical world, you now have a companion, a Cosmos 3 that can understand and reason. It can generate. It can simulate in the loop. It can even be the policy itself. It is on the top of leaderboards all over the all over the world. I am incredibly proud of Cosmos and today we're announcing Cosmos 3. Let's take a look.
## 01:41:01
**中文**
>> 现实世界无限而且不可预测。>> [音乐] >> 物理 AI 需要数据，但现实世界数据无法规模化。对于物理 AI，计算本身就是数据。>> [音乐] >> 这就是 Cosmos：面向物理 AI 的开放前沿全模态模型，采用新的混合 Transformer 架构。[音乐] 像素、动作、声音和语言进入自回归 Transformer，由它推理、规划并指挥扩散 Transformer 生成下一步内容。[音乐] 开发者可以针对不同机器人形态和用例对 Cosmos 进行后训练。
**English / Original**
>> The real world is infinite and unpredictable. >> [music] >> Physical AI needs data, but real-world data is impossible to scale. For physical AI, compute >> [music] >> is data. This is Cosmos, an open frontier omnimodel for physical AI built on a new mixture of transformers [music] architecture. Pixels, action, sound and language flow into the autoregressive [music] transformer which reasons, plans and instructs the diffusion transformer which generates what comes next. [music] Developers post-train Cosmos across embodiments and use cases.
## 01:41:39
**中文**
作为 VLM，Cosmos 观察物理世界，理解正在发生什么、描述场景并标出重要信息。[音乐] 作为世界模型，Cosmos 可从图像、文本或视频生成符合物理规律的合成视频；作为模拟器，它为策略训练与评估闭合循环。>> [音乐] >> 作为 NVIDIA Omniverse 的基础，Cosmos 还是一个以动作为条件的世界模型，能够逐帧预测未来。[音乐]
**English / Original**
As a VLM, Cosmos watches the physical world, understands [music] what's happening, describing scenes, and flagging what matters. As a world model, Cosmos [music] generates physics-accurate synthetic video from an image, text, or video. As a simulator, Cosmos closes the loop for policy training and evaluation. >> [music] >> And as the foundation of NVIDIA Omniverse, an action-conditioned world model, Cosmos predicts the future frame [music] by frame.
## 01:42:13
**中文**
训练完 Cosmos 后，它就成为了世界行动模型。为各种机器人、为[音乐]一切移动的事物感知、推理、规划、生成动作。一种新的数据，一种新的老师，通过计算生成[音乐]。 Cosmos，物理人工智能时代开发者的基础[音乐]。 >> [掌声]
**English / Original**
Post-train Cosmos, and it becomes a world action model. Perceiving, reasoning, planning, generating actions for robots of every kind, for [music] everything that moves. A new kind of data, a new kind of teacher, generated [music] by compute. Cosmos, the foundation [music] for developers of the age of physical AI. >> [applause]
## 01:42:55
**中文**
>> 这需要数据加上计算才能给你人工智能。现在我们有了人工智能，计算就是数据。因此，使用 Cosmos 3 训练一大堆 AI 模型。 Cosmos 是一个令人难以置信的开放模型系统。它与 NeMo Triton 完全相同。我们打开模型。我们开放数据，甚至开放我们如何训练它，以便您可以自己增强它，并将 Cosmos 变成您的专有模型。我们在许多不同的行业都有如此令人难以置信的合作伙伴与我们合作。
**English / Original**
>> This takes data plus compute gives you AI. Now that we have AI, compute is data. And so, use Cosmos 3, train a whole bunch of AI models. Cosmos is such an incredible open model system. It's exactly the same as NeMo Triton. We open model. We open the data, and we even open how we trained it so that you could enhance it for yourself and turn Cosmos into your proprietary model. We have such incredible partners working with us in so many different industries.
## 01:43:27
**中文**
现在，模型本身当然是 AI 堆栈中最容易理解的部分，但是 AI 堆栈非常复杂。它有生成器、模型、模拟器和运行时。就像智能体系统一样，这些汽车本质上是一个物理AI 智能体机器人，是一辆自动驾驶汽车，也有这个复杂的堆栈。
**English / Original**
Now, the model itself is the most, of course, the most understandable part of the AI stack, but the AI stack is very complicated. It has generators, the model, simulators, and the runtime. Just as Just as it is for agentic systems, these cars are essentially a physical AI agentic robot that is a is a autonomous vehicle has also this complicated stack.
## 01:43:56
**中文**
今天我们发布 Alpamayo 2，一款面向自动驾驶汽车的开放模型。我们正与全球汽车公司合作。看看这些采用 NVIDIA Hyperion、正在制造 NVIDIA Hyperion 汽车的品牌：其制造商合计约占全球汽车产量的 80%。未来会有大量 NVIDIA Hyperion 系统，可以运行 Alpamayo，也可以运行任何其他厂商的自动驾驶软件栈。
**English / Original**
Today, we're announcing AlphaMine 2, an open model for self-driving cars. We're working with car companies across the world. If you look at these brands that have signed up for the NVIDIA Hyperion, that are building NVIDIA Hyperion cars, this represents about 80% of the world's cars. The manufacturers represent 80% of the world's cars. We are going to have a whole lot of NVIDIA Hyperion systems that are able to run AlphaMine or anybody else's AV stack.
## 01:44:32
**中文**
我们还接入了出行服务。全球约 97% 的出行服务正在与我们连接；当我们把 Alpamayo 部署在采用 Halos 操作系统的 Hyperion 运行时上，就能连接世界各地的这些服务。来看一下。>> 嘿，Mercedes，带我去最喜欢的三明治店。
**English / Original**
We are also connected into mobility services. Approximately 97% of the world's mobility services are connecting with us so that when we deploy AlphaMine on the Hyperion runtime with the Halos operating system, we will be able to connect to all of these services across the world. Let's take a look at this. >> Hey Mercedes, let's go to my favorite sandwich shop.
## 01:44:59
**中文**
>> 前往目的地的路线。车道清晰。拉出开始行驶。由于前方静止的领先车辆挡住了我们的车道，请向左轻推[音乐]。减速停在控制[音乐]路口的停车标志处。因为行人在我们的车道上，所以停下来给行人让行。让路给从左边切入的车辆[音乐]。向左轻推以清除右侧阻挡的停止车辆。
**English / Original**
>> Routing to your destination. Lane is clear. Pulling out to start drive. Nudge left [music] due to the stationary lead vehicle ahead blocking our lane. Slow down to stop at the stop sign controlling [music] the intersection. Stop to yield to the pedestrian since the person is in our lane. Yield to the cut-in vehicle [music] from the left. Nudge left to clear the stopped vehicle blocking on the right.
## 01:45:22
**中文**
与切入车辆保持距离，因为它正在并入我们的车道。由于停止的货车挡住了我们车道右侧的[音乐]，请向左轻推。由于前方交叉而向右微移。停下来与我们车道正前方的领头车辆保持距离。与我们车道正前方的车辆保持距离[音乐]。如果路口有停车标志控制，请根据停车标志停车。由于车辆正在前方横穿马路，请让行。与领头车辆保持距离。由于卡车挡住了我们车道的右侧，请向左轻推。由于卡车挡住了我们车道的左侧，因此向右微移。
**English / Original**
Keep distance to the cut-in vehicle since it is merging into our lane. Nudge left due to the stopped van blocking [music] the right side of our lane. Nudge right due to crossing ahead. Stop to keep distance to the lead vehicle directly ahead in our lane. Keep distance [music] to the vehicle directly ahead in our lane. Stop for the stop signs if the intersection is controlled by a stop sign. Yield to cross traffic since the vehicle is crossing ahead. Keep distance to the lead vehicle. Nudge left due to the truck blocking the right side of our lane. Nudge right due to the truck blocking the left side of our lane.
## 01:45:43
**中文**
由于卡车挡住了我们车道的左侧，请向左轻推。保持距离 您的目的地位于右侧。 >> [音乐] >> 睁开我的眼睛。 >> [掌声] >> 世界上第一辆推理自动驾驶汽车。如果你一直让它说话，它会让你发疯的。但是，我们很高兴它一直在自言自语。
**English / Original**
Nudge left due to the truck blocking the left side of our lane. Keep distance Your destination is on the right. >> [music] >> Open my eyes. >> [applause] >> The world's first reasoning autonomous vehicle. If you let it talk all the time, it will drive you crazy. But, we're very happy that it's talking to itself all the time.
## 01:46:15
**中文**
这就是所谓的思考。所以，《睁开我的眼睛》是一辆推理车。我们创造的技术也适用于人形机器人。当然，还有许多新的突破必须发生。 NVIDIA Isaac Groot 是我们的人形机器人堆栈。模型数据生成模拟运行运行时包括操作系统。这代表了Groot平台，Isaac Groot平台。
**English / Original**
That's called thinking. And so, Open my eyes is a reasoning car. The technology that we've created also applies to humanoids. Of course, there many new breakthroughs that has to happen. The NVIDIA Isaac Groot is our humanoid robotic stack. model data generation simulation the runs the run time including the operating system. This represents Groot platform, the Isaac Groot platform.
## 01:46:48
**中文**
正如您所看到的，我们的每一个系统都具有完全相同的模式，无论是云智能体系统、PC 智能体系统、自动驾驶汽车的机器人系统、人形机器人的机器人系统，都是一样的。当然，在每一个案例中，我们都会完整地构建一切。我们垂直构建一切，与协同设计、极端协同设计完全集成，然后我们将其开放给每个人使用您喜欢的任何部分。
**English / Original**
Every one of our systems, as you can see, the exact same pattern, whether it's agentic system for the cloud, agentic system for the PC, a robotic system for a self-driving car, a robotic system for a humanoid robot, all the same. And of course, in every single case, we build everything completely. We build everything vertically, completely integrated with co-design, extreme co-design, and then we open it up for everybody to use whichever part you like.
## 01:47:22
**中文**
无论您想使用什么，我们甚至可以帮助您修改。但缺少的一件事是我们需要一个机器人系统的参考平台。这些机器人系统是如此复杂，有如此多的电机、如此多的传感器，如此脆弱。然而，我们需要有一种方法来提供这些参考平台，就像我们对 PC、DGX、云和自动驾驶汽车所做的那样，我们现在将为机器人做到这一点。今天，我们宣布推出 NVIDIA Isaac Groot，这是一款完全集成的参考人形机器人。 Sharpa 制造的每只手都有 25 个自由度。
**English / Original**
And whatever you want to use, we even help you modify. But the one thing that is missing is we need a reference platform for robotic systems. These robotic systems are so complicated, so many motors, so many sensors, so fragile. And yet we need to have a way to deliver these reference platforms just like we do with PCs and DGXs and clouds and self-driving cars, we now are going to do it for robots. Today we're announcing the NVIDIA Isaac Groot, a reference humanoid robot, all fully integrated. 25 degrees of freedom on the on each hand made by Sharpa.
## 01:48:02
**中文**
机器人全身有 31 个自由度，身高 6 英尺、体重 150 磅，就像我一样——第一个数字比我高，第二个数字比我小，除此之外很接近。这个平台运行新的 Thor 和我们的完整软件栈：数据生成栈、仿真栈和运行时，全部集成进一台人人可用的机器人。我们为高等教育和大学研究人员打造它，因为自行构建这一整套系统极其困难。下面来看一看。
**English / Original**
31 degrees of freedom on the robot. 6 ft, 150 lbs. Just like me. The first number is shorter, the second number is bigger. Otherwise, pretty close. And and this platform runs the new Thor and our entire software stack. Data generation stack, data simulation stack, the runtime, all integrated into a robot that is designed for everyone to use. Now, we built this for higher education and university researchers because for them to build this is insanely hard to do. And so let's take a
## 01:48:44
**中文**
看看那个。
**English / Original**
look at that.
## 01:48:46
**中文**
>> 人工智能[音乐]的下一个飞跃是通用机器人、类人机器人。但建立一个很难。每个团队都从头开始，将模拟器、远程操作系统、数据管道、[音乐]和培训基础设施拼接在一起。研究开始前需要进行数月的设置。 NVIDIA Isaac Groot，一个用于人形机器人的开放开发平台[音乐]。
**English / Original**
>> The next leap in AI [music] is general purpose robots, humanoids. But building one is hard. Every team starts from scratch, stitching together simulators, teleop systems, data pipelines, [music] and training infrastructure. Months of setup before research can start. NVIDIA Isaac Groot, an open development platform [music] for humanoid robots.
## 01:49:08
**中文**
开放模型、模拟和训练库以及数据生成器。再加上机器人[音乐]电脑。管道完全清洁，数小时内即可投入使用。首先，在Isaac Lab中搭建仿真环境。在真实或模拟机器人上捕捉 [音乐] Isaac Teleop 的演示。使用 Omniverse 和 Cosmos 生成合成数据，将一次演示扩展到数千次。
**English / Original**
Open models, simulation and training libraries, and data generators. Plus the robot [music] computer. Fully pipe clean, ready to go in hours. First, set up the simulation environment in Isaac Lab. Capture demonstrations with [music] Isaac Teleop on a real or simulated robot. Generate synthetic data with Omniverse and Cosmos, scaling one demonstration into thousands.
## 01:49:42
**中文**
训练策略，并在 Isaac Lab Arena 中评估；通过 Isaac ROS 部署，在 Jetson Thor 上运行。每个组成部分都是模块化、开放的：可以使用我们的，也可以替换成自己的。Groot 正在推动各个学科、各个领域的机器人研究，从研究实验室一直到工厂车间。[音乐] 一套开放平台。
**English / Original**
Train policies, evaluate them in Isaac Lab Arena. Deploy through Isaac ROS, running on Jetson Thor. Every element modular, open. Use ours or swap in your own. Groot is powering robotics research across every discipline, for every domain, from research labs to factory floors. [music] One open platform.
## 01:50:27
**中文**
现在，又增加了一个新成员。
**English / Original**
And now, a new addition.
## 01:50:30
**中文**
艾萨克·格鲁特参考设计机器人。基于 NVIDIA 的开放平台构建，可为任何地点的任何实验室进行前沿研究。机器人时代从这里开始[音乐]。NVIDIA艾萨克·格鲁特。 >> 这么多机器人。 >> [掌声] >> 我们正在与世界上几乎所有从事机器人或机器人系统工作的人合作。
**English / Original**
Isaac Groot reference design robots. Built on Nvidia's open platform, ready for frontier research for any lab, anywhere. The age of robotics starts [music] here. Nvidia Isaac Groot. >> So many robots. >> [applause] >> We're working with just about everybody who's working on robots in the world or robotic systems in the world.
## 01:51:01
**中文**
让我总结一下。计算机行业已经彻底改变：过去六个月里，一切都变了。原因是智能体得以实现，并与最新的前沿模型汇合，使 AI 如今可以完成有用的工作。这一计算模式将一次又一次重现：智能体由模型、调用带技能工具的编排框架以及运行时组成。运行时取决于它是在云端、本地、PC 还是机器人上运行，但它们的计算模式完全相同。
**English / Original**
Let me tell you what I told you. The computer industry has been completely changed. In the last 6 months, everything changed. Everything changed because agents were realized and it converged with the latest frontier models and it made possible the AI to now do useful work. The computing pattern will repeat over and over and over again. This computing pattern of an agent that's a model, a harness that uses tools with skills and runs in a runtime. That runtime depends on whether it's in the cloud or on prem, on a PC or a robot. But the computing pattern is exactly the same for all of
## 01:51:43
**中文**
它们都是如此。
**English / Original**
them.
## 01:51:44
**中文**
你会按偏好选择不同的编排框架和模型，并为专有用途改进它们；你还会创建超级子智能体，把它们租给其他人，帮助对方完成工作。针对这种智能体平台与计算模式，NVIDIA 提供了企业 AI 工具包。
**English / Original**
You will use different harnesses because of your preference. You will use different models because of your preference. You will improve them for your proprietary use. You would create sub super agents that you can rent to other people to help them do their work. This agentic platform, this agentic pattern Nvidia has an enterprise AI toolkit.
## 01:52:05
**中文**
对于所有人来说，这是一种与人工智能互动的绝佳方式，对于我们来说，这也是一个绝佳的成长机会。Vera Rubin 已全面投入生产。而 Grace Blackwell 的创建是为了处理人工智能，特别是推理。Vera Rubin 的诞生是为了管理智能体。目前已全面投入生产。它不仅仅是 GPU。它是整个分解的分布式智能体处理系统。
**English / Original**
This is a wonderful way for all of you to engage your AIs and for us it's a wonderful growth opportunity. Vera Rubin is in full production. Whereas Grace Blackwell was created to process AI, particularly inference. Vera Rubin was created to run agents. It is in full production. It is much much more than a GPU. It is entire disaggregated, distributed agent processing system.
## 01:52:35
**中文**
NVIDIA已经真正成为一家基础设施公司。不仅仅是一家 GPU 公司，不仅仅是一家系统公司，而是一家基础设施公司，可以帮助您产生最大的收入、最大的利润并尽快实现这一目标。智能体世界是一种新的计算方式，您现在为智能体而不是为人构建 CPU。
**English / Original**
Nvidia has really become an infrastructure company. Not just a GPU company, not just a systems company, but an infrastructure company to help you generate the maximum revenues, the maximum profit and to get there as soon as possible. The agent world this new way of doing computing where you build CPUs now for agents, not for people.
## 01:52:59
**中文**
智能体的 CPU 有其自身的特殊要求，而我们的 NVIDIA Vera 是革命性的。我对它的内存非常满意。该订单已经使其成为我们公司历史上最快、最成功的产品发布。NVIDIA和微软创建了全新的个人电脑系列。这是一个新的开始。当然，与我刚才描述的智能体处理模式、计算模式完全相同的智能体模式也将在各种设备上运行。我提到了个人电脑，但未来，它将是机器人、卫星和基站，
**English / Original**
CPUs for agents has its own special requirement and our Nvidia Vera is revolutionary. I'm so happy about its RAM. The order is already is going to make it the fastest and the most successful product launch in our company's history. Nvidia and Microsoft has created a whole new line of PCs. This is a new beginning. And of course, that exact same agentic pattern that I agentic processing pattern, computing pattern that I just described is also going to run on all kinds of devices. I mentioned PCs, but in the future, it'll be robots and satellites and base stations and
## 01:53:38
**中文**
工厂、云端、本地、边缘。这种模式，智能体 AI 系统，这种智能体计算模式将在全世界的计算机中复制。我们对个人电脑的看法很可能会改变。我要感谢你们所有人的伙伴关系和友谊。如果没有我们一起做的一切，我们就不可能来到这里。我为你们去年取得的如此成功感到非常自豪。
**English / Original**
factories, in the cloud, on prem, at the edge. This pattern, agentic AI system, this agentic computing pattern will be replicated in computers all over. How we think about the personal computer will very likely change. I want to thank all of you for your partnership, your friendship. We couldn't be here without everything that we do together. I am so proud of how you've been so successful this last year.
## 01:54:05
**中文**
明年将会更多。我还有一件事要告诉你。我们来看一下。 >> [尖叫] [音乐] >> 台湾，你准备好了吗？让我们这样做吧。 >> 主题演讲将在台北国际电脑展上举行。 [音乐] 黄仁勋 向世界展示了接下来会发生什么。有用的人工智能已经到来。智能体在您身边[音乐]工作。但万一你
**English / Original**
The next year is going to be even more. I have one more thing for you. Let's take a look. >> [screaming] [music] >> You ready, Taiwan? Let's do this. >> The keynote's coming at Computex. [music] Jensen showed the world what's next. Useful AI has arrived. Agents working by [music] your side. But in case you
## 01:54:48
**中文**
错过了我们今天所说的事情，我们将为你分解这一切，台北。 [音乐] >> 智能体曾经被误解。在好莱坞，只有电影明星才会拥有它们。现在我们都有团队[音乐]让梦想成真，在客厅里建立公司。但他们需要大量的计算。我们[音乐]听到你的声音。 >> 这就是我们创建 Vera 的原因。
**English / Original**
missed things we said today, we're going to break it all down for you, Taipei. [music] >> Agents used to be misunderstood. Only movie stars had them in Hollywood. Now we all got teams [music] making dreams come true, building companies from living rooms. But they need so much compute. We [music] hear you. >> That's why we created Vera.
## 01:55:07
**中文**
>> Rubin 定律，事实证明是真的：最低成本的 token 正源源而来。>> 推理速度提升 10 倍，智能体比 007 还多。[音乐] >> BlueField 可靠保存智能体记忆。>> 现在来谈 CPU。>> 快 50%，太夸张了。>> 对 Vera 来说并不夸张，它就是为智能体打造。>> NVLink Fusion 巧妙融合 ASIC，[音乐] 欢迎所有人加入 NVLink 阵营。
**English / Original**
>> Ruben's law, the show is true. The cheapest tokens coming through. >> 10 times faster inference heaven. More special agents [music] than 007. >> Bluefield keeps agents memory true. >> Now, let's talk about it, CPU. >> 50% faster, that's outrageous. >> Not for Vera. >> It's built for agents. >> Envion fusion blends [music] ASICs smartly. Everyone's welcome to the Envion party.
## 01:55:32
**中文**
>> 如果你喜欢刚才的开场：>> Vera Rubin，>> 已全面投产。>> Nemotron 3 Ultra 领跑，[音乐] 工作速度提升 5 倍。NeMo Guardrails 把护栏守好，OpenShell 把沙箱锁紧。[音乐] >> 你的代码已经迁移并完成审查，>> 在这首歌结束之前一切就已完成。[音乐] >> AI 是创造收入的五层蛋糕，这一点不要搞错。
**English / Original**
>> Well, if you like that introduction. >> Zero Ruben's. >> In full production. >> Nemo trials ultra leave the [music] run. Five x faster work gets done. Nemo clock keeps the guardrails right. Open shell keeps the sandbox [music] tight. >> Your code migrated and reviewed. >> All before this [music] song is through. >> AI is a five layer cake for revenue. Make no mistake.
## 01:55:57
**中文**
>> 全球人工智能[音乐]云拥有大量千兆瓦。接下来，Giga 保持电力精简，将点连接起来。 >> 每一瓦都为您优化。 >> 这样你就可以吃蛋糕了。 >> 也吃掉它。 >> RTX 标志着这一年。 40 年来最重要的 PC [音乐] 时刻。 >> 智能体为我们的工作流程提供动力。在 Windows 所在的任何地方运行。 >> 编排框架在 CPU 上运行。
**English / Original**
>> Global AI [music] clouds with lots of gigawatts. Giga next keeps power lean, connecting dots. >> Every watt optimized for you. >> So you can have your cake. >> And eat it, too. >> RTX marks the year. Biggest PC [music] moment in 40 years. >> Agents powering our workflows. Running anywhere [music] Windows goes. >> Harnesses run on CPU.
## 01:56:19
**中文**
>> 模型在 GPU 上飞驰。>> Cosmos 构建机器人所需的世界，[音乐] 把计算转化为合成数据。>> Alpamayo 看得见、能推理，像人一样理解交通流。[音乐] >> GR00T 让机器人学会运动，掌握技能、找到节奏。>> 还需要由 Thor 驱动的系统。[音乐] >> 但它仍然只是个人形机器人。
**English / Original**
>> Models fly on GPU. >> Cosmos build [music] worlds that robots need. >> Turning computing into synthetic feet. >> Alpha my OC's and reasons through. >> Understand [music] flows like people do. >> Who is how they learn to move. >> Learning skills and finding groove. >> You'll [music] need trees powered by those. >> It's still just humanoid.
## 01:56:41
**中文**
快点。 >> [音乐] [音乐] [歌唱] >> 来吧。 >> [音乐] [歌唱] >> 未来是光明的。快来看看接下来会发生什么吧。 >> 谢谢台湾。 >> [叹息与惊呼]
**English / Original**
Come on. >> [music] [music] [singing] >> Come on. >> [music] [singing] >> The future's bright. Come see what's next. >> Thank you, Taiwan. >> [sighs and gasps]
## 01:57:22
**中文**
>> 欢迎来到 Computex。>> [音乐] >> 大家谢谢。>> [掌声] >> 祝大家度过精彩的 Computex。大家谢谢。感谢你们带来非凡的一年。>> [掌声] >> 感谢大家的友谊与支持，谢谢。保重，祝大家度过精彩的 Computex。>> [掌声] [音乐]
**English / Original**
>> Welcome to Computex. >> [music] >> Dàjiā xièxie. >> [applause] >> HAVE A GREAT COMPUTEX. Dàjiā xièxie. Thanks for an amazing year. >> [applause] >> Thank you for all your friendship and support. Thank you. Take care. Have a great Computex. >> [applause] [music]