Nvidia Enters CPU and AI PC Markets with N1X Chip and Vera Rubin Platform at Computex 2026

InfraTop News11 sources·Jun 2

chips-and-hardware compute-infrastructure on-device-and-edge nvidia n1x vera-rubin

Summary

Updated Sep 10 — Jensen Huang reiterates 70% revenue growth forecast at Goldman Sachs, projecting ~$680B next fiscal year; Blackwell system sees 27% month-to-month sales growth.

• Nvidia announced the N1X / RTX Spark superchip for AI-focused Windows PCs, launching fall 2026
• Microsoft, Dell, and others will build PCs around the Arm-based chip, targeting professionals and gamers
• Vera Rubin platform introduces Nvidia's first-ever standalone CPU for data centers, entering Intel and AMD territory
• Nvidia holds $5T+ market cap after 450% stock gains over three years, backed by $215B+ in annual revenue

Adjust signal

Updates

Sep 10

Type	Key Point	Context
Financials	Huang reiterates 70% YoY revenue growth at Goldman Sachs; ~$400B→~$680B projected next fiscal year	At Goldman Sachs Communicopia + Technology conference on September 10, 2026, Huang said 'I think we could grow 70% year over year. We're confident about that.' Analysts expect Nvidia to end its current fiscal year at approximately $400 billion in revenue — 70% growth would project ~$680 billion for next year. The 70% guidance was first given last month when Nvidia reported yet another record-breaking quarter.
Stat	Blackwell GPU system (36 Grace CPUs + 72 Blackwell GPUs) seeing 27% month-to-month sales growth	Huang highlighted one product — an integrated compute system combining 36 Grace CPUs with 72 Blackwell GPUs — as currently experiencing 27% month-to-month sales growth, signaling accelerating demand for Nvidia's newest-generation hardware.
Tech Info	Single Blackwell 'GPU': $8.5M price, 2 million parts, 250,000 kilowatts; ships by airplane	Huang challenged the perception of Nvidia as a chip company: 'One GPU now is not $399. It's $8.5 million dollars... 2 million parts, right? 250,000 kilowatts. That's a GPU, and we ship thousands of them.' This illustrates how Nvidia's product has evolved from discrete consumer cards to multi-rack data center infrastructure.
Strategy	Circular investment defense: $100B in verified customer contracts before any Nvidia equity stake	Responding to concerns about 'circular deals' (investing in companies that then buy Nvidia chips), Huang cited $100 billion in visible third-party contracts that must exist before any investment: 'I'm not taking any risks... I need a sure thing.' He joked that putting '$1 in and $100 comes back' is not a risk — it's a deal worth doing more of.
Context	Competition from hyperscalers, AI labs (Anthropic, OpenAI), Cerebras, and Etched acknowledged	Huang acknowledged competition from Amazon, Microsoft, and Google (building custom AI chips), from AI labs building their own silicon, and from chip startups including recently-public Cerebras and Etched. He argued Nvidia maintains dominance because it 'runs every model' across all these customers and tracks every gigawatt of data center power and land globally.

Financials

Huang reiterates 70% YoY revenue growth at Goldman Sachs; ~$400B→~$680B projected next fiscal year

At Goldman Sachs Communicopia + Technology conference on September 10, 2026, Huang said 'I think we could grow 70% year over year. We're confident about that.' Analysts expect Nvidia to end its current fiscal year at approximately $400 billion in revenue — 70% growth would project ~$680 billion for next year. The 70% guidance was first given last month when Nvidia reported yet another record-breaking quarter.

Stat

Blackwell GPU system (36 Grace CPUs + 72 Blackwell GPUs) seeing 27% month-to-month sales growth

Huang highlighted one product — an integrated compute system combining 36 Grace CPUs with 72 Blackwell GPUs — as currently experiencing 27% month-to-month sales growth, signaling accelerating demand for Nvidia's newest-generation hardware.

Tech Info

Single Blackwell 'GPU': $8.5M price, 2 million parts, 250,000 kilowatts; ships by airplane

Huang challenged the perception of Nvidia as a chip company: 'One GPU now is not $399. It's $8.5 million dollars... 2 million parts, right? 250,000 kilowatts. That's a GPU, and we ship thousands of them.' This illustrates how Nvidia's product has evolved from discrete consumer cards to multi-rack data center infrastructure.

Strategy

Circular investment defense: $100B in verified customer contracts before any Nvidia equity stake

Responding to concerns about 'circular deals' (investing in companies that then buy Nvidia chips), Huang cited $100 billion in visible third-party contracts that must exist before any investment: 'I'm not taking any risks... I need a sure thing.' He joked that putting '$1 in and $100 comes back' is not a risk — it's a deal worth doing more of.

Context

Competition from hyperscalers, AI labs (Anthropic, OpenAI), Cerebras, and Etched acknowledged

Huang acknowledged competition from Amazon, Microsoft, and Google (building custom AI chips), from AI labs building their own silicon, and from chip startups including recently-public Cerebras and Etched. He argued Nvidia maintains dominance because it 'runs every model' across all these customers and tracks every gigawatt of data center power and land globally.

Details

Type	Key Point	Context
Product Launch	Nvidia unveiled the N1X chip inside its RTX Spark superchip for Windows PCs, launching fall 2026	The chip uses Arm architecture rather than x86, which Nvidia claims delivers higher performance and efficiency than existing Intel-based designs. Microsoft, Dell, and other OEMs will manufacture the initial devices. The 14mm form factor targets professionals, gamers, and users wanting ultra-portable machines.
New Tech	RTX Spark PCs are designed from the ground up to run AI agents natively	Jensen Huang framed the launch as Microsoft and Nvidia jointly reinventing the PC. The emphasis on AI agent execution signals this is not merely a performance upgrade but a platform shift aimed at AI workloads running locally on consumer and professional hardware.
Product Launch	Vera Rubin platform includes Nvidia's first-ever standalone CPU for data centers	Previously, Nvidia offered CPU components only as part of integrated systems. The standalone CPU marks Nvidia's formal entry as a first-class competitor in the server CPU market, directly challenging Intel Xeon and AMD EPYC product lines.
Market Impact	Nvidia is expanding from a GPU supplier into end-to-end AI silicon across both data centers and client devices	The dual announcements represent a deliberate broadening of Nvidia's addressable market. The company is no longer solely dependent on selling accelerators into cloud infrastructure; it is now competing across the full silicon stack, from edge devices to hyperscale servers.
Context	Nvidia reached $5T+ market cap and $215B+ in revenue heading into these announcements	The stock has gained roughly 450% over the past three years, driven almost entirely by AI infrastructure demand. That financial position gives Nvidia the capital and leverage to fund aggressive market expansion into CPU and PC segments.
Market Impact	The CPU market entry puts Nvidia in direct competition with Intel and AMD on their core home turf	Intel and AMD have historically owned both the data center CPU and consumer PC processor markets. Nvidia entering with Arm-based silicon introduces a well-capitalized, AI-credentialed competitor into a market both incumbents are already under pressure to defend.
Insight	Jensen Huang also pitched 'insane' AI ROI to billionaire investors at a closed-door Computex event	At a separate Taipei forum hosted by Era and Chailease Holding, Huang told 300+ financial institution and family office representatives that AI ROI has been 'completely reset' and is now 'insanely profitable,' directly rebutting investor skeptics who questioned AI spending returns.

Product Launch

Nvidia unveiled the N1X chip inside its RTX Spark superchip for Windows PCs, launching fall 2026

The chip uses Arm architecture rather than x86, which Nvidia claims delivers higher performance and efficiency than existing Intel-based designs. Microsoft, Dell, and other OEMs will manufacture the initial devices. The 14mm form factor targets professionals, gamers, and users wanting ultra-portable machines.

New Tech

RTX Spark PCs are designed from the ground up to run AI agents natively

Jensen Huang framed the launch as Microsoft and Nvidia jointly reinventing the PC. The emphasis on AI agent execution signals this is not merely a performance upgrade but a platform shift aimed at AI workloads running locally on consumer and professional hardware.

Product Launch

Vera Rubin platform includes Nvidia's first-ever standalone CPU for data centers

Previously, Nvidia offered CPU components only as part of integrated systems. The standalone CPU marks Nvidia's formal entry as a first-class competitor in the server CPU market, directly challenging Intel Xeon and AMD EPYC product lines.

Market Impact

Nvidia is expanding from a GPU supplier into end-to-end AI silicon across both data centers and client devices

The dual announcements represent a deliberate broadening of Nvidia's addressable market. The company is no longer solely dependent on selling accelerators into cloud infrastructure; it is now competing across the full silicon stack, from edge devices to hyperscale servers.

Context

Nvidia reached $5T+ market cap and $215B+ in revenue heading into these announcements

The stock has gained roughly 450% over the past three years, driven almost entirely by AI infrastructure demand. That financial position gives Nvidia the capital and leverage to fund aggressive market expansion into CPU and PC segments.

Market Impact

The CPU market entry puts Nvidia in direct competition with Intel and AMD on their core home turf

Intel and AMD have historically owned both the data center CPU and consumer PC processor markets. Nvidia entering with Arm-based silicon introduces a well-capitalized, AI-credentialed competitor into a market both incumbents are already under pressure to defend.

Insight

Jensen Huang also pitched 'insane' AI ROI to billionaire investors at a closed-door Computex event

At a separate Taipei forum hosted by Era and Chailease Holding, Huang told 300+ financial institution and family office representatives that AI ROI has been 'completely reset' and is now 'insanely profitable,' directly rebutting investor skeptics who questioned AI spending returns.

Product Launch = new hardware announced; New Tech = novel capability or architecture; Market Impact = competitive or financial consequence; Context = background data; Insight = analytical interpretation of events

What This Means

Nvidia is executing a deliberate expansion from a single-category GPU supplier into a full-stack AI silicon company, competing in consumer PCs, professional workstations, and data center CPUs simultaneously. For AI builders and deployers, this means local AI agent execution on client hardware becomes a realistic near-term deployment target, reducing reliance on cloud inference for certain workloads. Jensen Huang's aggressive investor messaging at Computex — declaring AI ROI "insanely profitable" — underscores Nvidia's intent to sustain the AI infrastructure investment cycle and fend off skepticism about AI spending returns. For investors and incumbents, the competitive threat to Intel and AMD from a company with Nvidia's capital base, AI ecosystem lock-in, and direct OEM partnerships with Microsoft and Dell is structurally significant.

Sentiment

Broadly excited with analysts highlighting Nvidia's aggressive expansion into CPUs and AI PCs

@jjfleagleJason Fleagle · Head of AI @ NetsyncView post

Impressed

“Nvidia is no longer just powering the AI data center. It is coming for the PC. At Computex, Nvidia unveiled RTX Spark, a new Windows AI PC platform built with Microsoft and supported by major OEMs like Dell, HP, ASUS, Lenovo, MSI, and Surface. The bigger story is the AI factory moving to the desk. RTX Spark combines Grace CPU, Blackwell GPU, and up to 128GB of unified memory to support local AI agents.”

@NeblinaTechNeblina Tech - Cloud Info · Cloud technology analystView post

Excited

“Nvidia acaba de declarar guerra a Intel, AMD y ARM al mismo tiempo. En Computex 2026, Jensen Huang presentó Vera Rubin — la primera CPU propia de Nvidia, basada en arquitectura ARM — y anunció que se asocia con Microsoft para lanzar los primeros PCs Windows con chips Nvidia. Lo llamó 'la mayor reinvención del PC en 40 años.'”

@leofinanzasLeonardo · Trader focused on financial marketsView post

Supportive

“$NVDA entra al mercado de PCs. Nvidia presentó ayer en Computex, Taiwán, el RTX Spark (N1X), su primer superchip diseñado para computadoras personales. Desarrollado en conjunto con MediaTek y orientado a correr Windows para Arm... El anuncio tuvo impacto inmediato en el mercado. Trepó más de 6% en la sesión, mientras Dell y HP subieron más de 10% y 8%, respectivamente. Intel cayó más del 4%.”

@peesamacpeesamac · AI workflow practitionerView post

Impressed

“NVIDIA เพิ่งทำสิ่งที่หลายคนรอมาสามปี เมื่อวาน (1 มิ.ย.) ที่งาน Computex 2026 เจนเซ่น หวง เปิดตัว RTX Spark Superchip (โค้ดเนม N1X) ซึ่งเป็นชิป CPU ตัวแรกในประวัติศาสตร์ของ NVIDIA ที่ลงมาเล่นตลาด PC กับโน้ตบุ๊กเต็มตัว... มันมาในจังหวะที่ชนกับ Apple M-series แบบตรง ๆ”

@optionslabproOptionsLabPro · Options trading analystView post

Supportive

“The catalyst is Jensen Huang's COMPUTEX keynote, where Nvidia unveiled the N1X, its first system-on-chip for Windows laptops. The N1X pairs a 20-core Arm-based CPU with an RTX-class Blackwell GPU. Pair that with the Vera Rubin platform, also Arm-based, entering full production with roughly 3.5 times the training performance of Blackwell. Arm is now the architecture behind both the next Nvidia data-center CPU and the next Nvidia laptop SoC.”

Split

~80/20 positive/negative — dominant view celebrates Nvidia's full-stack move and market impact; small minority notes competitive pressure on Intel/AMD without deeper criticism.

Sources

Update history (2)

Sep 10Jensen Huang reiterates 70% revenue growth forecast at Goldman Sachs, projecting ~$680B next fiscal year; Blackwell system sees 27% month-to-month sales growth.Added Jensen Huang Goldman Sachs Communicopia conference coverage (Sept 10, 2026): 70% revenue growth reiteration projecting ~$680B next fiscal year, Blackwell system 27% month-to-month growth, $8.5M GPU system details, circular investment defense citing $100B in verified contracts, and competition context.

Jun 3Added Jensen Huang's closed-door investor pitch at Computex 2026 Taipei forum. New details: Huang declared AI ROI "completely reset" and "insanely profitable" to 300+ representatives from family offices and financial institutions (Hillhouse, PAG, DBS). Added as new tier3_deep_dive row and incorporated into what_this_means and key_facts. New article linked from Bloomberg/Yahoo Finance.

Similar Events

Nvidia GTC 2026: Jensen Huang's Bullish Keynote Fails to Move Wall Street

Mar 21

Nvidia Launches Open-Source Quantum AI Models to Make Quantum Computing Practical

May 3