NVIDIA has unveiled a near-zero-water cooling system for AI data centres. Using a closed-loop liquid cooling design, the technology recirculates coolant instead of consuming water, reducing energy use, operating costs and dependence on traditional cooling towers.

NVIDIA has unveiled a near-zero-water cooling system for AI data centres. Using a closed-loop liquid cooling design, the technology recirculates coolant instead of consuming water, reducing energy use, operating costs and dependence on traditional cooling towers.

Can AI data centres run without water? Nvidia introduces near-zero-water cooling technology

Microsoft and Nvidia have introduced a powerful new category of personal computers designed to handle tasks previously reserved for large server rooms or desktop workstations. It handles 120-billion parameter models locally.

Microsoft and Nvidia have introduced a powerful new category of personal computers designed to handle tasks previously reserved for large server rooms or desktop workstations. It handles 120-billion parameter models locally.

This one computer can now replace your entire office!

Nvidia is overtaking traditional chipmakers by powering the Microsoft Surface Laptop Ultra with its RTX Spark superchip. Featuring up to 128GB of unified memory and 1 petaflop of AI compute, the chip allows users to run massive 120-billion parameter models locally.


Nvidia is overtaking traditional chipmakers by powering the Microsoft Surface Laptop Ultra with its RTX Spark superchip. Featuring up to 128GB of unified memory and 1 petaflop of AI compute, the chip allows users to run massive 120-billion parameter models locally.

‘Nvidia vs traditional chipmakers’: How Nvidia is winning the race that traditional chipmakers started

Microsoft and Nvidia have unveiled the Surface Laptop Ultra, an innovative PC powered by the new RTX Spark superchip that delivers 1 petaflop of AI compute and 128GB of unified memory.

Microsoft and Nvidia have unveiled the Surface Laptop Ultra, an innovative PC powered by the new RTX Spark superchip that delivers 1 petaflop of AI compute and 128GB of unified memory.

Is this the beginning of the Nvidia-powered PC era?

Microsoft has introduced the Surface Laptop Ultra, a new premium laptop developed in partnership with Nvidia that the company describes as its most powerful Surface device to date. The announcement marks a significant step in Microsoft's AI PC strategy as the company looks to bring advanced artificial intelligence capabilities directly onto personal computers.

<blockquote class="twitter-tweet">Introducing Surface Laptop Ultra. Built for world makers. Designed for what&#39;s next. The most powerful Surface laptop ever. Coming Fall 2026. Sign up to learn more: <a href="https://t.co/k8aEX2pTAy">https://t.co/k8aEX2pTAy</a> <a href="https://t.co/bY1NdzNnie">pic.twitter.com/bY1NdzNnie</a>&mdash; Microsoft Surface (@surface) <a href="https://x.com/surface/status/2061329225886453989?ref_src=twsrc%5Etfw">June 1, 2026</a></blockquote> <script async src="https://platform.x.com/widgets.js" charset="utf-8"></script>

The new device is powered by Nvidia's RTX Spark processor, an Arm-based chip designed to handle demanding AI workloads while maintaining the portability and battery life expected from modern laptops. Microsoft says the Surface Laptop Ultra represents years of collaboration between the two companies on Windows on Arm and AI-focused computing.<h2>What is the Surface Laptop Ultra?</h2>The Surface Laptop Ultra is a 15-inch laptop built around Nvidia's new RTX Spark platform. It is the first major Surface device to combine Microsoft's Windows ecosystem with Nvidia's latest Arm-based AI computing architecture. The launch is particularly notable because it marks Microsoft's renewed effort to build premium Arm-powered computers. Earlier attempts faced challenges, but advances in AI processing, chip design and software optimisation have created new opportunities for Arm-based Windows devices.Microsoft says the laptop has been designed for creators, developers, AI researchers and professionals who need high-performance computing in a portable form factor.RTX Spark: The chip at the centre of Microsoft's AI pushThe biggest talking point is Nvidia's RTX Spark processor.According to Microsoft and Nvidia, the chip can feature:Up to 20 CPU coresUp to 6,144 GPU coresUp to 128GB unified memoryAI performance approaching 1 petaflopDedicated AI acceleration capabilitiesThe architecture is based on technology previously used in Nvidia's DGX Spark systems for AI developers. It has now been adapted for Windows devices, allowing advanced AI applications to run locally on the machine rather than relying entirely on cloud infrastructure. Industry observers see this as part of a wider trend where AI workloads are increasingly moving closer to the user rather than being processed exclusively in remote data centres.<h2>Designed for AI, creators and developers</h2>Microsoft says the Surface Laptop Ultra has been built for users handling demanding workloads such as AI model development, video editing, software development, 3D design and content creation. The company claims graphics performance is comparable to laptops equipped with Nvidia RTX 5070-class hardware, while AI processing capabilities are designed to support increasingly complex local AI applications. As businesses and consumers adopt more AI-powered tools, hardware manufacturers are focusing on devices capable of running these workloads efficiently without sacrificing portability.<h2>Display, design and battery life</h2>The Surface Laptop Ultra features a 15-inch mini-LED display with a pixel density of 262 pixels per inch. Microsoft says it is the brightest display ever used in a Surface device, with peak HDR brightness reaching 2,000 nits. The company has also introduced its largest Surface trackpad to date, complete with haptic feedback technology. The laptop weighs around 2 kilograms and will be available in silver and dark grey finishes.Microsoft also says the device is designed to deliver all-day battery life, although detailed battery performance figures have not yet been released.<h2>Ports and connectivity</h2> Unlike many premium laptops that have reduced port options, the Surface Laptop Ultra includes a broad selection of connectivity features.The company has confirmed support for:USB-C USB-A HDMI Full-size SD card slot 3.5mm headphone jackMicrosoft has not yet disclosed the full technical specifications for these ports, including transfer speeds and supported standards.<h2>Why this launch matters</h2>The launch is important not only because of the hardware itself but also because it highlights a deeper partnership between Microsoft and Nvidia. The two companies have spent years working on Windows on Arm and AI-focused computing. The Surface Laptop Ultra is one of the first major products to emerge from that collaboration. It also reflects a growing industry shift towards AI PCs. Companies including Microsoft, Apple, Qualcomm, AMD and Intel are all investing heavily in devices designed specifically for artificial intelligence workloads.Rather than focusing only on traditional performance metrics, manufacturers are increasingly highlighting AI processing power, memory bandwidth and on-device AI capabilities.<h2>More RTX Spark devices are expected</h2>Microsoft has indicated that the Surface Laptop Ultra will not be the only product based on Nvidia's RTX Spark platform. According to the companies, the architecture is expected to appear in additional laptops and compact desktop systems in the future. This suggests that RTX Spark could become a broader ecosystem rather than a single flagship device.As AI becomes a larger part of everyday computing, Microsoft and Nvidia appear to be positioning themselves at the centre of the next generation of personal computers. The Surface Laptop Ultra is expected to launch later in 2026, although Microsoft has not yet announced pricing or exact availability dates. For now, the announcement provides a glimpse into how Microsoft sees the future of computing: powerful AI capabilities running directly on devices, supported by specialised hardware and deep integration between software and silicon. Whether the Surface Laptop Ultra can reshape the Windows laptop market remains to be seen, but it clearly signals Microsoft's determination to compete aggressively in the emerging AI PC era.&nbsp;

Microsoft has introduced the Surface Laptop Ultra, a new premium laptop developed in partnership with Nvidia that the company describes as its most powerful Surface device to date. 

Microsoft has introduced the Surface Laptop Ultra, a new premium laptop developed in partnership with Nvidia that the company describes as its most powerful Surface device to date. The announcement marks a significant step in Microsoft's AI PC strategy as the company looks to bring advanced artificial intelligence capabilities directly onto personal computers.

Microsoft has introduced the Surface Laptop Ultra, a new premium laptop developed in partnership with Nvidia that the company describes as its most powerful Surface device to date.&nbsp;

Microsoft unveils Surface Laptop Ultra with Nvidia RTX Spark chip: Key things to know

Nvidia CEO and founder Jensen Huang used his keynote address at GTC Taipei 2026 to outline the company's vision for the next phase of artificial intelligence.The presentation focused on AI infrastructure, new computing architectures, agentic AI systems and Nvidia's growing partnerships with major technology companies. Huang argued that artificial intelligence is moving beyond chatbots and simple assistants towards systems capable of reasoning, planning and carrying out complex tasks.Here are the five biggest announcements and takeaways from Nvidia's latest AI event.<h2>Nvidia and Microsoft deepen AI partnership</h2>One of the most significant announcements from the event was Nvidia's expanded partnership with Microsoft. According to Huang, the two companies are working together on a new generation of AI-powered personal computers. These systems will reportedly use Nvidia's RTX Spark superchip and run a version of Windows designed to support agentic AI experiences. The move highlights Microsoft's growing focus on AI-native computing and Nvidia's ambition to bring advanced AI capabilities directly to personal computers. Industry analysts believe partnerships like this could intensify competition across the PC market, where companies are increasingly racing to integrate AI capabilities into future devices.<h2>Agentic AI takes centre stage</h2>A major theme throughout Huang's presentation was agentic AI. Unlike traditional AI assistants that mainly respond to prompts, agentic AI systems are designed to complete multi-step tasks, make decisions and work more independently. Huang said the industry is now entering a phase where AI can reason through complex problems and assist users in more meaningful ways. According to Nvidia, many of the products announced during the keynote are built around these capabilities. Technology companies across the industry are increasingly investing in agentic AI, with many experts viewing it as the next major stage of artificial intelligence development.<h2>Nvidia says AI is becoming a business tool</h2>Huang also focused on the economic impact of artificial intelligence. He argued that AI is no longer simply an experimental technology and is increasingly becoming a practical tool that companies can use to improve productivity and generate returns on investment. Businesses across sectors are adopting AI for software development, customer service, research, automation and operational planning. Nvidia believes the growing use of AI agents could further increase these gains over the coming years. The comments reflect a broader shift in the AI industry, where attention is moving from experimentation towards measurable business outcomes.<h2>Introducing Vera, Nvidia's new CPU architecture</h2>Another major announcement was Vera, Nvidia's latest CPU architecture. According to the company, Vera has been designed specifically for AI workloads rather than traditional consumer computing tasks. The architecture is powered by Nvidia's custom Olympus CPU core and features 88 cores along with high-bandwidth memory capabilities. Nvidia says these systems are intended to support AI-driven computing environments where intelligent agents handle many of the tasks that would traditionally require human interaction. The announcement demonstrates how semiconductor companies are increasingly designing specialised hardware tailored specifically for artificial intelligence applications.<h2>Vera Rubin becomes Nvidia's next AI platform</h2>Huang also highlighted Vera Rubin, Nvidia's next-generation AI infrastructure platform. The system is expected to succeed the company's Blackwell architecture and is designed to support large-scale AI workloads used by enterprises, cloud providers and AI research organisations. Vera Rubin is built for the growing demand for AI training, inference and agentic computing systems. The company expects the platform to power future data centres and AI infrastructure projects around the world.Reports suggest that major AI companies, including Anthropic, OpenAI and SpaceXAI, are among the organisations exploring next-generation Nvidia infrastructure solutions.<h2>What do these announcements mean for the AI industry?</h2>The keynote highlighted a broader trend shaping the technology sector. The conversation is increasingly moving beyond individual AI models and towards the infrastructure needed to support them. Companies are now investing heavily in specialised chips, AI data centres, agentic systems and next-generation computing platforms.Nvidia remains at the centre of that transformation. The company's processors power many of the world's leading AI systems, and its technology continues to play a critical role in the expansion of artificial intelligence. For investors, businesses and developers, Huang's message was clear: the next phase of AI will depend not only on smarter models but also on the computing infrastructure capable of supporting them at scale.

Nvidia CEO and founder Jensen Huang used his keynote address at GTC Taipei 2026 to outline the company's vision for the next phase of artificial intelligence.


Nvidia CEO and founder Jensen Huang used his keynote address at GTC Taipei 2026 to outline the company's vision for the next phase of artificial intelligence. The presentation focused on AI infrastructure, new computing architectures, agentic AI systems and Nvidia's growing partnerships with major technology companies. Huang argued that artificial intelligence is moving beyond chatbots and simple assistants towards systems capable of reasoning, planning and carrying out complex tasks. Here are the five biggest announcements and takeaways from Nvidia's latest AI event. Nvidia and Microsoft deepen AI partnership One of the most significant announcements from the event was Nvidia's expanded partnership with Microsoft. According to Huang, the two companies are working together on a new generation of AI-powered personal computers. These systems will reportedly use Nvidia's RTX Spark superchip and run a version of Windows designed to support agentic AI experiences. The move highlights Microsoft's growing focus on AI-native computing and Nvidia's ambition to bring advanced AI capabilities directly to personal computers. Industry analysts believe partnerships like this could intensify competition across the PC market, where companies are increasingly racing to integrate AI capabilities into future devices. Agentic AI takes centre stage A major theme throughout Huang's presentation was agentic AI. Unlike traditional AI assistants that mainly respond to prompts, agentic AI systems are designed to complete multi-step tasks, make decisions and work more independently. Huang said the industry is now entering a phase where AI can reason through complex problems and assist users in more meaningful ways. According to Nvidia, many of the products announced during the keynote are built around these capabilities. Technology companies across the industry are increasingly investing in agentic AI, with many experts viewing it as the next major stage of artificial intelligence development. Nvidia says AI is becoming a business tool Huang also focused on the economic impact of artificial intelligence. He argued that AI is no longer simply an experimental technology and is increasingly becoming a practical tool that companies can use to improve productivity and generate returns on investment. Businesses across sectors are adopting AI for software development, customer service, research, automation and operational planning. Nvidia believes the growing use of AI agents could further increase these gains over the coming years. The comments reflect a broader shift in the AI industry, where attention is moving from experimentation towards measurable business outcomes. Introducing Vera, Nvidia's new CPU architecture Another major announcement was Vera, Nvidia's latest CPU architecture. According to the company, Vera has been designed specifically for AI workloads rather than traditional consumer computing tasks. The architecture is powered by Nvidia's custom Olympus CPU core and features 88 cores along with high-bandwidth memory capabilities. Nvidia says these systems are intended to support AI-driven computing environments where intelligent agents handle many of the tasks that would traditionally require human interaction. The announcement demonstrates how semiconductor companies are increasingly designing specialised hardware tailored specifically for artificial intelligence applications. Vera Rubin becomes Nvidia's next AI platform Huang also highlighted Vera Rubin, Nvidia's next-generation AI infrastructure platform. The system is expected to succeed the company's Blackwell architecture and is designed to support large-scale AI workloads used by enterprises, cloud providers and AI research organisations. Vera Rubin is built for the growing demand for AI training, inference and agentic computing systems. The company expects the platform to power future data centres and AI infrastructure projects around the world. Reports suggest that major AI companies, including Anthropic, OpenAI and SpaceXAI, are among the organisations exploring next-generation Nvidia infrastructure solutions. What do these announcements mean for the AI industry? The keynote highlighted a broader trend shaping the technology sector. The conversation is increasingly moving beyond individual AI models and towards the infrastructure needed to support them. Companies are now investing heavily in specialised chips, AI data centres, agentic systems and next-generation computing platforms. Nvidia remains at the centre of that transformation. The company's processors power many of the world's leading AI systems, and its technology continues to play a critical role in the expansion of artificial intelligence. For investors, businesses and developers, Huang's message was clear: the next phase of AI will depend not only on smarter models but also on the computing infrastructure capable of supporting them at scale.

Nvidia CEO and founder Jensen Huang used his keynote address at GTC Taipei 2026 to outline the company's vision for the next phase of artificial intelligence.

‘Nvidia’s next AI push’: 5 biggest announcements from CEO Jensen Huang’s GTC Taipei 2026 keynote

Nvidia CEO Jensen Huang has gone viral after videos showed him dancing alongside employees during a company event in Taipei, Taiwan.The clips, which have been widely shared across social media platforms, show Huang joining staff members on stage during Nvidia's Constellation All-Employee Celebration. The videos quickly attracted attention online, with many users praising the technology executive's relaxed and approachable personality. The moment comes at a time when Nvidia remains one of the most closely watched companies in the global artificial intelligence industry, making Huang one of the most recognised business leaders in the world.

<blockquote class="twitter-tweet" data-media-max-width="560">🇺🇸Jensen Huang is out there living his best life, dancing at the NVIDIA all-employee celebration event in Taipei, Taiwan You&#39;d never guess he&#39;s the CEO 😂<a href="https://t.co/LsLTsMfPqd">pic.twitter.com/LsLTsMfPqd</a>&mdash; Mario Nawfal (@MarioNawfal) <a href="https://x.com/MarioNawfal/status/2060683137324126491?ref_src=twsrc%5Etfw">May 30, 2026</a></blockquote> <script async src="https://platform.x.com/widgets.js" charset="utf-8"></script>

<h2>What happened at the Nvidia event?</h2>During the employee celebration in Taipei, Huang stepped onto the stage and joined a dance performance with Nvidia staff members. Videos from the event show the Nvidia chief matching dance moves with employees while interacting with the audience. The performance drew applause and cheers from attendees and soon spread across social media platforms including X and Instagram. Many viewers were surprised to see the leader of one of the world's most valuable technology companies taking part in such a light-hearted moment.The clips quickly generated millions of views and became one of the most discussed moments from Nvidia's activities in Taiwan this week.<h2>Why are people talking about the video?</h2>Much of the online reaction focused on Huang's leadership style and his relationship with employees. Many users said the video reflected a workplace culture where senior leadership remains accessible and engaged with staff. Others pointed to the contrast between Huang's role as the head of a company worth trillions of dollars and his willingness to participate in informal employee activities.Comments on social media described Huang as "down-to-earth", while others said the video showed a more personal side of one of the technology industry's most influential figures. The reaction also reflects Huang's growing popularity beyond the technology sector. Over the past two years, his public profile has risen significantly alongside Nvidia's success in artificial intelligence and advanced computing.<h2>How Jensen Huang became one of tech's most recognisable leaders</h2>Huang co-founded NVIDIA in 1993 and has led the company for more than three decades. Under his leadership, Nvidia transformed from a graphics chip manufacturer into one of the most important companies powering the global AI boom. The company's processors are widely used for training and running advanced artificial intelligence models. As demand for AI computing increased, Nvidia's market value surged, making it one of the world's most valuable publicly traded companies. Unlike many technology executives, Huang has also become known for his public appearances, interactions with customers and regular engagement with employees.The viral dance video is not the first time Huang has attracted attention outside the boardroom During previous visits to Asia, he has been seen participating in cultural activities, visiting local markets and interacting with the public. In Taiwan, he has frequently been spotted at popular night markets and restaurants, where crowds often gather to take photographs and speak with him.These appearances have helped build his reputation as one of the more approachable leaders in the technology industry. Industry observers say Huang's public image has become an important part of Nvidia's brand as the company continues expanding its influence across AI, cloud computing and data centre technologies. Huang is currently in Taiwan attending major artificial intelligence events and meetings with industry leaders. According to local reports, he is expected to travel to South Korea next week, where he will meet executives from companies including SK Group, Hyundai Motor Group, LG Corp and Naver. Discussions are expected to focus on AI infrastructure, robotics, semiconductor technology and the future of physical AI systems. These meetings come as countries and companies across Asia continue investing heavily in artificial intelligence and next-generation computing technologies.

Nvidia CEO Jensen Huang has gone viral after videos showed him dancing alongside employees during a company event in Taipei, Taiwan.


Nvidia CEO Jensen Huang has gone viral after videos showed him dancing alongside employees during a company event in Taipei, Taiwan. The clips, which have been widely shared across social media platforms, show Huang joining staff members on stage during Nvidia's Constellation All-Employee Celebration. The videos quickly attracted attention online, with many users praising the technology executive's relaxed and approachable personality. The moment comes at a time when Nvidia remains one of the most closely watched companies in the global artificial intelligence industry, making Huang one of the most recognised business leaders in the world.

Nvidia CEO Jensen Huang has gone viral after videos showed him dancing alongside employees during a company event in Taipei, Taiwan.

Nvidia CEO Jensen Huang goes viral after dancing with employees in Taiwan

Nvidia has been known for its powerful graphics processing units (GPUs). These chips have become central to artificial intelligence. However, the company is now building much more than chips, and this shift is becoming increasingly important.Under CEO Jensen Huang, Nvidia has expanded into networking, infrastructure and full-stack AI systems that power modern data centres.<h2>Nvidia’s second-biggest business: Networking</h2>Nvidia’s networking division has quietly grown into one of its largest revenue drivers.According to the company’s latest earnings:$11 billion revenue in a single quarter267 per cent year-on-year growthOver $31 billion annual revenueThis makes networking Nvidia’s second-largest segment after compute. Analysts say this growth is directly linked to rising demand for AI infrastructure.Kevin Cook, a strategist at Zacks Investment Research, noted that Nvidia’s quarterly networking revenue is now comparable to what some traditional networking companies generate in a full year.<h2>What exactly is Nvidia building?</h2>Nvidia’s networking business focuses on connecting thousands of GPUs inside data centres.Key technologies include:NVLink: connects GPUs within serversInfiniBand switches: enable high-speed data transferSpectrum-X: Ethernet platform designed for AI workloadsPhotonics switches: improve speed and energy efficiencyThese tools ensure that AI systems can process massive amounts of data quickly and efficiently.Instead of working as separate parts, these technologies are designed to function together as a single system.<h2>The idea of an ‘AI factory’</h2>Nvidia describes modern data centres as “AI factories”. These are facilities designed to train and run AI models at scale.In this setup:GPUs act as the computing engineNetworking acts as the backboneStorage and software complete the systemNvidia’s strategy is to provide all these components as an integrated package.This approach allows companies to build large-scale AI systems more efficiently.Mellanox acquisition: The turning pointNvidia’s expansion into networking began with its acquisition of Mellanox in 2020 for $7 billion. Mellanox, an Israeli company, specialised in high-performance networking.At the time, the move did not attract much attention. However, it has since become a key part of Nvidia’s growth.Industry experts say the acquisition helped Nvidia turn its GPUs into a complete solution by adding the missing networking layer.<h2>Full-stack strategy: Why it matters</h2>Unlike companies that sell individual components, Nvidia is focusing on a full-stack approach.This means it offers:Hardware (GPUs and networking)Software toolsIntegrated systemsThe company also works through partners to deploy these systems globally.According to Nvidia executives, this integrated model gives the company an advantage, as AI workloads require tight coordination between compute and networking.<h2>New announcements at GTC</h2>At its recent GTC conference, Nvidia announced several updates to strengthen its AI infrastructure:New Rubin platform for AI supercomputingImproved Spectrum-X networking systemsInference memory storage solutionsThese developments aim to make AI systems faster and more efficient.Why this matters for the AI raceThe global AI race is no longer just about building better models. It is also about building the infrastructure that powers them.<h2>AI models require:</h2>Massive computing powerHigh-speed data transferEfficient system integrationNvidia’s networking business addresses these needs.Experts say the company’s success lies in understanding that networking is not just a support function it is central to how AI systems operate.<h2>The bigger picture</h2>Nvidia’s shift shows how the AI industry is evolving. Companies are moving from standalone tools to complete ecosystems. While GPUs remain critical, networking and infrastructure are becoming equally important. As demand for AI continues to grow, the ability to build and manage large-scale systems may define the next phase of the industry. Nvidia’s strategy suggests that the future of computing is not just about chips it is about how everything connects.

NVIDIA is evolving beyond GPUs into a full-stack AI infrastructure company, with its networking business emerging as a major growth driver. The division generated $11 billion in quarterly revenue and is now Nvidia’s second-largest segment. 

Nvidia has been known for its powerful graphics processing units (GPUs). These chips have become central to artificial intelligence. However, the company is now building much more than chips, and this shift is becoming increasingly important. Under CEO Jensen Huang, Nvidia has expanded into networking, infrastructure and full-stack AI systems that power modern data centres. Nvidia’s second-biggest business: Networking Nvidia’s networking division has quietly grown into one of its largest revenue drivers. According to the company’s latest earnings: $11 billion revenue in a single quarter 267 per cent year-on-year growth Over $31 billion annual revenue This makes networking Nvidia’s second-largest segment after compute. Analysts say this growth is directly linked to rising demand for AI infrastructure. Kevin Cook, a strategist at Zacks Investment Research, noted that Nvidia’s quarterly networking revenue is now comparable to what some traditional networking companies generate in a full year. What exactly is Nvidia building? Nvidia’s networking business focuses on connecting thousands of GPUs inside data centres. Key technologies include: NVLink: connects GPUs within servers InfiniBand switches: enable high-speed data transfer Spectrum-X: Ethernet platform designed for AI workloads Photonics switches: improve speed and energy efficiency These tools ensure that AI systems can process massive amounts of data quickly and efficiently. Instead of working as separate parts, these technologies are designed to function together as a single system. The idea of an ‘AI factory’ Nvidia describes modern data centres as “AI factories”. These are facilities designed to train and run AI models at scale. In this setup: GPUs act as the computing engine Networking acts as the backbone Storage and software complete the system Nvidia’s strategy is to provide all these components as an integrated package. This approach allows companies to build large-scale AI systems more efficiently. Mellanox acquisition: The turning point Nvidia’s expansion into networking began with its acquisition of Mellanox in 2020 for $7 billion. Mellanox, an Israeli company, specialised in high-performance networking. At the time, the move did not attract much attention. However, it has since become a key part of Nvidia’s growth. Industry experts say the acquisition helped Nvidia turn its GPUs into a complete solution by adding the missing networking layer. Full-stack strategy: Why it matters Unlike companies that sell individual components, Nvidia is focusing on a full-stack approach. This means it offers: Hardware (GPUs and networking) Software tools Integrated systems The company also works through partners to deploy these systems globally. According to Nvidia executives, this integrated model gives the company an advantage, as AI workloads require tight coordination between compute and networking. New announcements at GTC At its recent GTC conference, Nvidia announced several updates to strengthen its AI infrastructure: New Rubin platform for AI supercomputing Improved Spectrum-X networking systems Inference memory storage solutions These developments aim to make AI systems faster and more efficient. Why this matters for the AI race The global AI race is no longer just about building better models. It is also about building the infrastructure that powers them. AI models require: Massive computing power High-speed data transfer Efficient system integration Nvidia’s networking business addresses these needs. Experts say the company’s success lies in understanding that networking is not just a support function it is central to how AI systems operate. The bigger picture Nvidia’s shift shows how the AI industry is evolving. Companies are moving from standalone tools to complete ecosystems. While GPUs remain critical, networking and infrastructure are becoming equally important. As demand for AI continues to grow, the ability to build and manage large-scale systems may define the next phase of the industry. Nvidia’s strategy suggests that the future of computing is not just about chips it is about how everything connects.

NVIDIA is evolving beyond GPUs into a full-stack AI infrastructure company, with its networking business emerging as a major growth driver. The division generated $11 billion in quarterly revenue and is now Nvidia’s second-largest segment.&nbsp;

What is Nvidia building beyond chips, and why does it matter now?

Nvidia has reportedly reached its largest-ever deal, a $20 billion licensing and talent acquisition agreement with AI chip startup Groq (not to be confused with the AI bot Grok from Elon Musk’s xAI). The deal smells like an attempt to gobble up a potential competitor, part of the “catch and kill” moves Silicon Valley’s big tech industry is notorious for. What is the deal, and why does it matter for one of the world’s top AI chipmaking giants? Here is what you should know.<h2>What is in the Groq deal with Nvidia?</h2>Started by former Google engineers in 2016, Groq makes LPU (Language Processing Unit) chips for ultra-fast, low-cost AI inference. The reported $20 billion all-cash deal includes asset purchases and non-exclusive technology licensing. These assets include Groq’s AI inference technology and intellectual property.Groq’s founder and CEO Jonathan Ross, president Sunny Madra, and other senior leaders will now be part of Nvidia talent force. This deal, if confirmed, surpasses Nvidia’s $7 billion acquisition of Mellanox in 2019. As of its September 2025 funding round, Groq was valued at $6.9 billion. GroqCloud, the cloud operations arm of the startup, is not part of the deal and will continue independently under new leadership, with CFO Simon Edwards likely to be its CEO, as per reports.Also read: <a href="https://www.wionews.com/world/nvidia-scripts-history-becomes-first-firm-to-reach-5-trillion-market-value-following-ai-boom-1761748694795">Riding on AI boom, Nvidia scripts history, becomes first firm to reach $5 trillion market value</a><h2>Why Groq matters for Nvidia</h2>AI inference is the process of running trained AI models, such as chatbots and other real-time applications, to generate instant responses. Currently, AI inference is becoming a larger share of total AI computing demand as models move from training to deployment at scale.&nbsp;This possibly explains Nvidia’s growing emphasis on inference optimisation alongside its dominance in AI training.Therefore, the Groq deal could bolster AI inference leadership for Nvidia, and eventually help integrate low-latency technology into its AI chips.<a href="https://www.wionews.com/videos/nvidia-hits-5tn-market-cap-world-business-watch-wion-1761801804595/embed">https://www.wionews.com/videos/nvidia-hits-5tn-market-cap-world-business-watch-wion-1761801804595/embed</a><h2>Why Groq’s tech is a coveted asset for Nvidia: The case for LPU chips</h2>Groq is led by ex-Googlers like Jonathan Ross, who had helped design Google’s Tensor Processing Unit (TPU). TPUs are specialised AI chips that accelerate complex AI and Machine Learning tasks behind Google's Search, Photos, Translate, and Gemini AI interface. It is also used in Google data centres, Google Cloud, devices and recommendation engines. &nbsp;Groq builds specialised AI accelerator chips called Language Processing Units (LPUs), designed specifically for fast, low-latency, and energy-efficient AI inference.While traditional graphics processing units (GPUs) are good at training AI models, Groq’s LPUs focus on deterministic, high-speed inference with lower power consumption.This makes them well-suited to heavier workloads that need quick token generation.Also read: <a href="https://www.wionews.com/photos/what-would-happen-to-nvidia-if-china-attacked-taiwan-today-a-factual-breakdown-of-the-worst-case-scenario-1764335291497">What would happen to NVIDIA if China attacked Taiwan today? A factual breakdown of the worst case scenario</a><h2>Catch and kill? What Nvidia is saying about Groq deal</h2>Reports citing Nvidia CEO Jensen Huang said the deal’s goal is to integrate Groq’s low-latency processors into Nvidia’s AI architecture to enhance inference capabilities. He insisted that Nvidia is not acquiring Groq as a company.So, the deal is perhaps an “acqui-hire” or “reverse acqui-hire”, similar to, but not exactly the same as, Microsoft’s arrangement with Inflection AI or Meta’s deal with Scale AI.Such deals allow access to talent, IPs, and technology while avoiding the regulatory scrutiny that would have occurred in the case of a full acquisition. Note that regulators are scrutinising partial acquisitions, asset purchases, and talent transfers in AI, as firms use these structures to avoid reviews of full mergers.At least for now, the Groq deal looks different from a catch-and-kill, or the ‘killer acquisition’ policy in which promising startups are acquired mainly to eliminate competition and shelve their technology rather than developing them further.Even then, one must note that Groq was indeed a direct competitor to Nvidia in AI inference, where the latter faced more rivals than in AI training.Nvidia, for now, says it is actively planning to integrate and scale Groq’s LPU technology into its own products, including future AI factories and real-time workloads. Huang’s comments highlight an expansion of capabilities, not suppression of talent or technology, even as the deal would end Groq as an independent rival and consolidate Nvidia’s dominance in inference.

Nvidia’s reported $20bn Groq deal targets the chip startup's AI inference chip technology. By licensing its LPU tech and hiring talent, Nvidia would strengthens low-latency AI computing. But is it a catch and kill approach?

Nvidia has reportedly reached its largest-ever deal, a $20 billion licensing and talent acquisition agreement with AI chip startup Groq (not to be confused with the AI bot Grok from Elon Musk’s xAI). The deal smells like an attempt to gobble up a potential competitor, part of the “catch and kill” moves Silicon Valley’s big tech industry is notorious for. What is the deal, and why does it matter for one of the world’s top AI chipmaking giants? Here is what you should know. What is in the Groq deal with Nvidia? Started by former Google engineers in 2016, Groq makes LPU (Language Processing Unit) chips for ultra-fast, low-cost AI inference. The reported $20 billion all-cash deal includes asset purchases and non-exclusive technology licensing. These assets include Groq’s AI inference technology and intellectual property. Groq’s founder and CEO Jonathan Ross, president Sunny Madra, and other senior leaders will now be part of Nvidia talent force. This deal, if confirmed, surpasses Nvidia’s $7 billion acquisition of Mellanox in 2019. As of its September 2025 funding round, Groq was valued at $6.9 billion. GroqCloud, the cloud operations arm of the startup, is not part of the deal and will continue independently under new leadership, with CFO Simon Edwards likely to be its CEO, as per reports. Also read: Riding on AI boom, Nvidia scripts history, becomes first firm to reach $5 trillion market value Why Groq matters for Nvidia AI inference is the process of running trained AI models, such as chatbots and other real-time applications, to generate instant responses. Currently, AI inference is becoming a larger share of total AI computing demand as models move from training to deployment at scale. This possibly explains Nvidia’s growing emphasis on inference optimisation alongside its dominance in AI training. Therefore, the Groq deal could bolster AI inference leadership for Nvidia, and eventually help integrate low-latency technology into its AI chips. https://www.wionews.com/videos/nvidia-hits-5tn-market-cap-world-business-watch-wion-1761801804595/embed Why Groq’s tech is a coveted asset for Nvidia: The case for LPU chips Groq is led by ex-Googlers like Jonathan Ross, who had helped design Google’s Tensor Processing Unit (TPU). TPUs are specialised AI chips that accelerate complex AI and Machine Learning tasks behind Google's Search, Photos, Translate, and Gemini AI interface. It is also used in Google data centres, Google Cloud, devices and recommendation engines. Groq builds specialised AI accelerator chips called Language Processing Units (LPUs), designed specifically for fast, low-latency, and energy-efficient AI inference. While traditional graphics processing units (GPUs) are good at training AI models, Groq’s LPUs focus on deterministic, high-speed inference with lower power consumption. This makes them well-suited to heavier workloads that need quick token generation. Also read: What would happen to NVIDIA if China attacked Taiwan today? A factual breakdown of the worst case scenario Catch and kill? What Nvidia is saying about Groq deal Reports citing Nvidia CEO Jensen Huang said the deal’s goal is to integrate Groq’s low-latency processors into Nvidia’s AI architecture to enhance inference capabilities. He insisted that Nvidia is not acquiring Groq as a company. So, the deal is perhaps an “acqui-hire” or “reverse acqui-hire”, similar to, but not exactly the same as, Microsoft’s arrangement with Inflection AI or Meta’s deal with Scale AI. Such deals allow access to talent, IPs, and technology while avoiding the regulatory scrutiny that would have occurred in the case of a full acquisition. Note that regulators are scrutinising partial acquisitions, asset purchases, and talent transfers in AI, as firms use these structures to avoid reviews of full mergers. At least for now, the Groq deal looks different from a catch-and-kill, or the ‘killer acquisition’ policy in which promising startups are acquired mainly to eliminate competition and shelve their technology rather than developing them further. Even then, one must note that Groq was indeed a direct competitor to Nvidia in AI inference, where the latter faced more rivals than in AI training. Nvidia, for now, says it is actively planning to integrate and scale Groq’s LPU technology into its own products, including future AI factories and real-time workloads. Huang’s comments highlight an expansion of capabilities, not suppression of talent or technology, even as the deal would end Groq as an independent rival and consolidate Nvidia’s dominance in inference.

Nvidia’s reported $20bn Groq deal targets the chip startup's AI inference chip technology. By licensing its LPU tech and hiring talent, Nvidia would strengthens low-latency AI computing. But is it a catch and kill approach?

Catch and kill? Why Nvidia’s record $20 billion deal with AI chip startup Groq matters

nvidia-ai-chip

Nvidia Ai Chip, News on Nvidia Ai Chip, stories on Nvidia Ai Chip, articles on Nvidia Ai Chip, Nvidia Ai Chip stories at Wion, Nvidia Ai Chip news, Nvidia Ai Chip latest news

nvidia ai-chip

Can AI data centres run without water? Nvidia introduces near-zero-water cooling technology

This one computer can now replace your entire office!

How Nvidia is winning the race that traditional chipmakers started

Is this the beginning of the Nvidia-powered PC era?

Microsoft launches most powerful Surface laptop yet with RTX Spark superchip

5 biggest announcements from CEO Jensen Huang’s GTC Taipei 2026 keynote

Nvidia CEO Jensen Huang goes viral after dancing with employees in Taiwan

Nvidia’s networking business surges as AI infrastructure demand grows

Why Nvidia’s record $20 billion deal with AI chip startup Groq matters