---
格式版本: 2
标题: "First look at Nvidia’s AI system Vera Rubin and how it beats Blackwell"
原文链接: "https://www.cnbc.com/2026/02/25/first-look-at-nvidias-ai-system-vera-rubin-and-how-it-beats-blackwell.html"
发布日期: "2026-02-25"
发现时间: "2026-02-25T00:00:00+08:00"
入库时间: "2026-05-28T02:39:38+08:00"
来源平台: "历史资料迁移"
搜索渠道: "legacy_migration"
搜索词: "NVIDIA Vera Rubin rack scale AI"
匹配关键词:
  - "AI Rack"
  - "Scale-up"
  - "NVIDIA Vera Rubin rack scale AI"
  - "待定数据"
  - "厂家"
  - "Nvidia"
  - "First"
  - "look"
  - "at"
  - "Nvidia’s"
  - "AI"
  - "system"
  - "Vera"
  - "Rubin"
  - "and"
  - "how"
  - "it"
  - "beats"
  - "Blackwell"
  - "25.md"
相关厂家:
  - "NVIDIA"
  - "AMD"
  - "Meta"
  - "Microsoft"
  - "Google"
  - "Foxconn"
  - "Broadcom"
相关专家:
  []
内容类型: "网页"
抓取工具: "Firecrawl 原始抓取 + qwen3.6-flash 原文清洗"
清洗工具: "历史资料原文保留 + LLM 正文裁剪"
关联判断: "待人工确认"
关联置信度: 0
关联理由: "历史资料迁移，未重新调用模型判定"
AI优质: "是"
AI打分: 91
AI分档: "高置信优质"
AI质检状态: "通过"
AI打分理由: "独家首探NVIDIA下一代机柜级AI系统Vera Rubin，详述72 GPU/36 CPU架构、18计算托盘、全液冷、模块化设计及功耗能效比。含明确量产时间、客户名单、供应链与定价预估，信息完整且具高商业与技术价值。"
AI质检模型: "qwen3.6-plus"
AI质检时间: "2026-06-13T07:00:15+08:00"
AI主题相关性: 20
AI来源权威性: 13
AI新颖性: 19
AI技术细节: 16
AI商业部署信号: 14
AI完整性: 9
采集批次: "2026年5月28日2点39分33秒"
采集批次ID: "legacy-migration-20260528023933"
去重键: "https://www.cnbc.com/2026/02/25/first-look-at-nvidias-ai-system-vera-rubin-and-how-it-beats-blackwell.html"
---

# Nvidia’s new AI system Vera Rubin is 10 times more efficient than its predecessor — here’s a first look
Key Points
- CNBC got an exclusive first look at Vera Rubin, Nvidia’s next AI system that’s due to ship in the second half of the year.
- Vera Rubin will use about twice as much power as Blackwell but will be far more efficient by delivering 10 times more performance per watt, Nvidia told CNBC.
- Nvidia dominates the artificial intelligence chip market but faces intensifying competition from companies including Advanced Micro Devices, Broadcom and Google.
![First look at Nvidia's Vera Rubin AI system — 1.3 million components and 10 times more efficient](https://image.cnbcfm.com/api/v1/image/108269471-1771975189715-Dion_Katie_Vera_Rubin1.jpg?v=1771975367&w=750&h=422&vtcrop=y)
[Nvidia’s](https://www.cnbc.com/quotes/NVDA/) earnings on Wednesday are expected to show booming sales of the company’s current rack-scale system. But all eyes are on its next AI system, Vera Rubin, which is scheduled to roll out later this year. Vera Rubin, which is made up of 1.3 million components, will deliver 10 times more performance per watt than its predecessor, Grace Blackwell, the company claims. That’s a significant development when energy consumption is one of the most critical issues facing the artificial intelligence build-out. CNBC got an exclusive first look at Vera Rubin at Nvidia’s headquarters in Santa Clara, California. Nvidia says the new AI system is a complex web of parts sourced from around the world. Its core chips include 72 Rubin graphics processing units, or GPUs, and 36 Vera central processing units, or CPUs, primarily made by [Taiwan Semiconductor Manufacturing Co](https://www.cnbc.com/quotes/TSM/). The other parts, from liquid cooling elements to power systems and compute trays, come from more than 80 suppliers in at least 20 countries, including China, Vietnam, Thailand, Mexico, Israel and the U.S. One big challenge the company faces is the soaring costs of memory due to a global [shortage](https://www.cnbc.com/2026/01/10/micron-ai-memory-shortage-hbm-nvidia-samsung.html) from all the AI-driven demand. Dion Harris, Nvidia’s AI infrastructure head, said in an interview that the company has been giving suppliers “very detailed forecasts.” “We’re aligning to make sure that everything we’re shipping will be met by our supply chain,” he said. “We’re in good shape.” It’s a critical moment for Nvidia, which dominates the market for AI processors but faces intensifying competition from [Advanced Micro Devices](https://www.cnbc.com/quotes/AMD/) as well as custom silicon from [Broadcom](https://www.cnbc.com/quotes/AVGO/) and [Google’s](https://www.cnbc.com/quotes/GOOGL/) homegrown tensor processing units. Nvidia has plans to manufacture up to $500 billion of AI infrastructure in the U.S. through 2029, including making Blackwell GPUs at [TSMC’s new Arizona fabs](https://www.cnbc.com/2024/12/13/inside-tsmcs-new-chip-fab-where-apple-will-make-chips-in-the-us-.html). Grace Blackwell went into production in 2024, and changed the game on how much compute was possible with a single system. Vera Rubin, which is expected to ship in the second half of 2026, takes the company to another level. Nvidia CEO Jensen Huang announced in January that the system was in full production. ![Nvidia founder and CEO Jensen Huang speaks about the Vera Rubin AI platform during a question and answer session with reporters at the annual Consumer Electronics Show (CES) in Las Vegas, Nevada on January 6, 2026. (Photo by Patrick T. Fallon / AFP via Getty Images)](./assets/02-image.cnbcfm.com.jpg) Nvidia founder and CEO Jensen Huang speaks about the Vera Rubin AI platform at the annual Consumer Electronics Show in Las Vegas, Jan. 6, 2026. Patrick T. Fallon | Afp | Getty Images “These things are massive. They’ve got all the compute, all the networking, all the cabling, all the cooling,” said Daniel Newman of research firm Futurum Group. “They’ve got all these systems pulled together into a single rack built for absolute greatest efficiency and greatest performance. And that’s just not how servers were historically built.” Last week, [Meta announced plans to use Vera Rubin](https://www.cnbc.com/2026/02/17/meta-nvidia-deal-ai-data-center-chips.html) in its data centers by 2027. Nvidia’s list of other expected Vera Rubin customers includes [OpenAI](https://www.cnbc.com/2025/06/10/openai-cnbc-disruptor-50.html), [Anthropic](https://www.cnbc.com/2025/06/10/anthropic-cnbc-disruptor-50.html), [Amazon](https://www.cnbc.com/quotes/AMZN/), Google and [Microsoft](https://www.cnbc.com/quotes/MSFT/). The racks, manufactured in the U.S. and elsewhere, including in Taiwan and at a new [Foxconn](https://www.cnbc.com/quotes/) plant in Mexico, weigh nearly 2 tons and have about 1,300 total microchips, compared with Grace Blackwell’s 864. Vera Rubin is a simpler, modular system intended to ease installs and repairs. Each superchip slides out of one of the rack’s 18 compute trays in seconds. In the Blackwell system, those components are soldered to the board. Nvidia said the new system will consume about twice as much power as its predecessor, but will be far more efficient because of that 10 times return on performance per watt. Jordan Klein, an analyst at Mizuho Securities, said what “matters the most” is “how many tokens per power consumed can you get.” “The more you can tweak that or move up the curve, the higher the return would be on the dollar you spend,” Klein said. ![Nvidia showed CNBC the Vera Rubin superchip, with two Rubin GPUs and one Vera CPU, and a total of 17,000 components, at Nvidia headquarters in Santa Clara, California, on February 13, 2026.](./assets/03-image.cnbcfm.com.jpg) Nvidia showed CNBC the Vera Rubin superchip, with two Rubin GPUs and one Vera CPU, and a total of 17,000 components, at Nvidia headquarters in Santa Clara, California. Vera Rubin is also Nvidia’s first system that’s 100% liquid cooled, which Harris said helps data centers consume “much less water” than traditional evaporative cooling. Nvidia doesn’t share rack pricing, but Futurum Group estimates the price will increase about 25% from Grace Blackwell, bringing the system price to somewhere around $3.5 million to $4 million. As major customers seek to diversify their reliance on the chipmaker, many are also filling AI servers with their own in-house silicon. CNBC [visited an Amazon Web Services data center](https://www.cnbc.com/video/2025/10/29/amazons-biggest-ai-data-center-comes-online-powering-anthropic.html) in October packed with “ultra-servers” made up of the company’s Trainium 2 chips. Meanwhile, Google’s data centers are loaded with racks of [its TPUs](https://www.cnbc.com/2024/08/23/how-google-makes-custom-cloud-chips-that-power-apple-ai-and-gemini.html). Later this year, Nvidia will see some big competition when rival AMD ships its first rack-scale system called Helios. The chipmaker just secured a [major commitment from Meta](https://www.cnbc.com/2026/02/24/meta-to-use-6gw-of-amd-gpus-days-after-expanded-nvidia-ai-chipdeal.html) for up to 6 gigawatts of capacity. “You’re going to see a lot of uptake because customers want more capacity, but they also want a viable second source to keep Nvidia honest,” Klein said. With respect to the competition, Harris said: “Hats off to anyone who’s going to try. But this is certainly not a simple endeavor.“ [Watch the video](https://www.cnbc.com/video/2026/02/25/deconstructing-nvidias-vera-rubin-the-predecessor-to-grace-blackwell.html) for a detailed look at Vera Rubin and how the system works.