Nvidia's Vera CPU aims for $200B market as company pursues dominance in agentic AI infrastructure

Jensen Huang says he's found a 'brand new' $200B market for Nvidia | TechCrunch

Nvidia founder and CEO Jensen Huang is, perhaps, one of the greatest corporate hype men of all time when it comes to his company. He may even surpass Salesforce's Marc Benioff when it comes to relentless optimism in his company's future and revenues. Even so, he delivers on the hype, quarter after quarter. Instead of cautioning you to view the proclamation that he's found a "brand new $200 billion TAM for Nvidia" with skepticism, I'd argue he's earned a bit of trust. Huang positioned this massive new market at the feet of Nvidia's new CPU product, Vera, which was introduced in March. Speaking on Wednesday's earnings call -- after Nvidia posted another record-breaking quarter with $81.6 billion in revenue and forecast $91 billion for the next -- Huang pitched Vera as a potentially transformative product. And one that already has promising sales figures. But no matter how well Nvidia delivers, Wall Street harbors anxiety over what will knock Nvidia from its perch. Lately, such fears have centered on the CPU. Nvidia is the king of the GPU, whereas historically the CPU markets were owned by companies like Intel and AMD. (Nvidia has made CPUs previously, of course, but that's not its core business.) For example, last month Amazon Web Services crowed about a giant contract it signed with Meta for millions of Amazon's homegrown AI CPUs. Amazon CEO Andy Jassy has been clear that he thinks AWS can do AI chips, both GPUs and CPUs, at least as well, and possibly better than Nvidia. But now, with the Vera CPU, which is sold alone and bundled with its Rubin GPU, Huang believes he's unlocked "a major new growth driver" for his company because Vera is, he believes, "the world's first CPU, purpose-built for agentic AI," Huang said on the call. "Vera opens a brand new $200 billion TAM for Nvidia, a market we have never addressed before, and every major hyperscaler and system maker is partnering with us to deploy it. The world is rebuilding computing for agentic AI and robotic physical AI. Nvidia sits at the center of these transitions," hype man Huang said. He explained that while the "thinking" part of an AI model uses GPUs, agents mostly run on CPUs. They use CPUs to do their assigned tasks and will, he predicts, run their own form of CPU-driven PCs. Vera is for agents because it's specifically designed to process tokens as fast as possible. This is opposed to classic cloud architecture CPUs designed with "cores," or the ability to run multiple instances of apps as fast as possible. That sounds logical, but with the major cloud providers as well as startups pursuing AI chip development, what makes him think that Nvidia will be the go-to source for agentic CPUs? Because, Huang says, Nvidia has already sold $20 billion worth of standalone Vera CPUs this year and we're only at the beginning. "The world has a billion users, human users. My sense is that the world is going to have billions of agents, not today. I mean, we're going to grow into it, but we'll have billions of agents, and those billions of agents will all use tools. And those tools are going to be like PCs, just like us humans using using PCs today," he said.

Analyst says Nvidia poised to capture two-thirds of the x86 server CPU market from Intel and AMD with expected $20 billion in revenue -- 'Nvidia is already on track' to deliver 4 million Vera CPUs in FY2027

"Vera CPU opens a brand-new $200 billion TAM for Nvidia, a market we have never addressed before," Cress said during the company's conference call with financial analysts and investors. "Every major hyperscale and system maker is partnering with us to get it deployed. We have visibility to nearly $20 billion in total CPU revenue this year, setting us up to become the world-leading CPU supplier." Nvidia later clarified that the $20 billion figure includes sales of Grace and Vera processors within Superchip combinations, NVL72 systems, and standalone CPUs sold either as racks aimed at agentic AI workloads or other applications. Intel's data center and AI (DCAI) division's revenue totaled $16.8 billion last year, whereas AMD's data center unit earned $16.635 billion in 2025. While CPUs account for the lion's share of earnings of these business units, their sales are by far not 100% of their revenue, so actual sales of Xeon and EPYC products are well below $16 billion. The entire x86 server CPU market is worth around $30 billion. Therefore, the $20 billion figure would indeed approach two-thirds of the traditional server CPU market, making Nvidia the world's No. 1 server CPU supplier. What makes Nvidia's statement especially remarkable is that while Grace CPUs are widely available and have shipped in huge quantities, its 88-core Vera CPU hasn't yet shipped in high volume. Furthermore, Nvidia has never meaningfully participated in mainstream server CPUs before. Given that Nvidia is poised to sell millions of Rubin data center GPUs, and every two of them will be attached to a Vera CPU, the company is almost guaranteed to sell plenty of CPUs. "We will sell millions of Rubin GPUs and every two of them is connected to a Vera [CPU]," Jensen Huang, chief executive of Nvidia, told analysts and investors. "Vera is used in [four] ways. The first is Vera Rubin [platform containing two Rubin GPUs and one Vera CPU]. The second use case is Vera as a standalone CPU. The third is Vera with CX9 and its software stack for storage. The fourth is Vera with CX9 alongside a software stack for security, compute isolation, and confidential computing." Conquering the CPU kingdom Given the dominance of x86 servers, alongside AMD's EPYC and Intel's Xeon CPUs in particular, it is hard to imagine that another company can outsell these highly popular products. Yet, it is more than possible, given the fact that Nvidia can price its CPUs well above the average selling prices (ASPs) of other x86 offerings, and still manage to outsell competitors. This is because Nvidia sells vertically integrated platforms rather than standalone CPUs or GPUs. While Nvidia is not 'known' for its CPUs, it is definitely not a new entrant. "Nvidia is in a unique situation, and I do not think we can really call them a 'new entrant," McCarron told Tom's Hardware Premium. Based on leaked estimates from Morgan Stanley Research, Nvidia will charge its hyperscale clients around $5,000 per Vera CPU when they purchase VR200 NVL72 machines later this year. Assuming that the estimate is correct, then selling CPUs worth $20 billion will require Nvidia to sell 4 million Vera units. Four million units is perfectly achievable for Nvidia, McCarron believes. "As far as delivering 4 million CPUs per year, Nvidia is already on track to deliver a number very near that for its GB300 and Rubin systems in FY2027 (so roughly Q2 2026 - Q1 2027 calendar year)," McCarron said. "So, their comment indicates some moderate upside to CPU shipments." In fact, Nvidia could probably sell considerably more than four million CPUs this fiscal year if it wanted to, and if it secured or reallocated additional capacity for production of its 88-core Arm-based processor, according to McCarron. "This really just comes down to customer demand and pricing/revenue allocation to get to the $20 billion," the analyst told us. "I would expect that the number is going to be heavily weighted towards the end of their fiscal year." Speaking of capacity, Nvidia has commitments for capacity and inventory of around $145 billion, so capacity allocation may not be a problem for the company. "We remain front-footed in securing sufficient supply to support our customers' growth," Cress said. "In Q1, we increased total supply, inclusive of inventory, purchase commitments, and prepaids, to $145 billion." AMD and Intel shipped nearly 20 million EPYC and Xeon SP processors for data center systems in 2025, Dean McCarron told us. Meanwhile, AMD's EPYC average selling price was about $1,325, whereas the ASP of Intel's Xeon SP was about $1,125, according to Mercury Research. That said, if Nvidia sells 4 million processors worth $20 billion, then it will not only outsell both AMD and Intel, but will become a formidable rival for both companies from a pure unit sales point of view, too.

Nvidia Exec Hand-Delivers First Vera CPU Systems to AI Heavy Hitters

Nvidia had a special delivery for OpenAI, SpaceX, Anthropic, and Oracle Cloud recently: the first Vera CPU systems. Nvidia's Ian Buck, VP of hyperscale and performance computing, "hand-delivered the first-ever Nvidia Vera CPUs to our partners at Anthropic, OpenAI, SpaceX, and Oracle Cloud," Nvidia tweeted on Monday. "Vera is Nvidia's first custom CPU, purpose-built for the age of agentic AI. This is just the beginning. The road to Vera-powered systems starts here. Thank you to our partners for being on this journey with us. The best is yet to come." "Agentic AI is creating a new CPU moment in the AI factory -- as models move from answering to acting, Vera is purpose-built to keep that work moving at scale," Buck said in a statement. Vera is built from 88 Nvidia Olympus cores, each of which Nvidia claims can deliver up to 50% greater performance than the Grace CPU cores that Nvidia has used in its current-generation Blackwell GPU systems. Buck made the first delivery of a Vera CPU system to Anthropic, stopping off at its SoMa offices in San Francisco. Then it was off to OpenAI's Mission Bay headquarters, just a short 10-minute drive away, to hand over another Vera system to OpenAI's head of compute infrastructure, Sachi Katti, before an hour-long trip to SpaceX's xAI offices in Palo Alto. There, Buck met up with SpaceX CEO Elon Musk. SpaceX is reportedly considering Vera for building new reinforcement learning workloads into its training stack. The final outfit receiving the personal treatment from Nvidia was Oracle Cloud Infrastructure, which has already pledged to deploy hundreds of thousands of Nvidia Vera CPUs in 2026. Traveling to the Oracle AI Customer Excellence Center in Sydney, Australia, Buck met with product management lead Karan Batta and chief customer and partner success officer Gary Miller. Buck suggested Oracle will be able to offer customers "production-grade agentic AI infrastructure," at a scale no other cloud provider can match. Considering the Customer Excellence Center is designed to court Oracle AI customers, Vera will no doubt be a key exhibit for visitors. But Vera is only one-half of the ultimate Nvidia equation in 2026. As well as standalone deployments, Vera is the host processor for Nvidia's Vera Rubin NVL72 stack, which combines Nvidia's NVLink and twin Rubin GPUs together for a comprehensive, scalable AI solution. Those CPU and GPU systems are expected to go on sale later this year, completing another Nvidia annual hardware refresh. For gamers who have seen Nvidia and AMD release new chips at a near-annual cadence, I do wonder whether these companies are ready for how quickly this kit can become obsolete. For now, though, Nvidia is hyping its next CPU and further driving the AI FOMO that has been so prevalent over the past few years.

Nvidia on track to be worlds leading CPU supplier claims CFO

Already the planet's largest supplier of GPUs, Nvidia now intends to conquer the CPU market. "We have visibility to nearly $20 billion in total CPU revenue this year, setting us up to become the world's leading CPU supplier," Nvidia CFO Colette Kress said during the company's Q1 2027 earnings call on Wednesday. Nvidia is no stranger to CPUs having announced its first Arm datacenter chip, codenamed Grace, back in 2021. However until recently the company integrated most of these parts into GPU systems that users almost always deployed in AI datacenters and supercomputers. That changed in February when Nvidia revealed Meta was among the first hyperscalers now deploying standalone Grace CPU Superchips in its datacenters to power a variety of workloads including the Social Network's AI agents. At its GTC conference in March, Nvidia officially expanded its CPU line up to include a standalone Vera CPU system. Each chip features 88 custom Olympus Arm cores with support for simultaneous multi-threading (SMT) -- that's Hyperthreading in Intel speak -- along with confidential computing capabilities. Nvidia can equi[ each chip with up to 1.5 TB of LPDDR5x SOCAMM memory, which offers higher memory bandwidth at up to 1.2 TB/s and uses little power (which is why it's often used in laptops). "Vera will deliver up to 1.5x faster performance per core, 2x performance per watt, and 4x density per rack compared to x86-based alternatives," Kress claimed. Nvidia's reference designs pack up to two Vera CPUs onto a single board and via high-speed NVLink interconnects. Nvidia's Vera is also paired in a 2:1 ratio of Rubin GPUs to CPUs in its most powerful rack-scale AI compute platforms. Since the chip was detailed this spring, Kress claims nearly every major hyperscaler and system builder plans to deploy the chips. This week, several top AI labs and hyperscalers, including Anthropic, OpenAI, Oracle, and SpaceX took delivery of Nvidia's first Vera-based systems. "Vera CPU opens a brand new $200 billion TAM for Nvidia, a market we have never addressed before," she said. While Nvidia is expanding its addressable market to include standalone CPUs, it should be noted that much like the company's Ethernet networking products, they're designed primarily with AI and HPC applications in mind. The chips can't replace x86 processors in every application, yet. Kress' comments come as Nvidia caps off a strong end to the first quarter of its 2027 fiscal year. The GPU giant raked in $58.3 billion in profits on $81.6 billion in revenue for the quarter, the latter of which grew 85 percent YoY and 20 percent from the prior quarter. Kress attributed the sequential jump to an "inflection in inference demand." The quarter saw Nvidia change how it breaks out revenues. The company's business units have now been organized into a datacenter group which includes cloud, hyperscale, neocloud and enterprise sales, plus an edge group, which serves as a catchall for gaming, robotics, automotive, and vRAN products. Datacenter revenues accounted for the vast majority of revenues, at $75.2 billion. Of that $38 billion came from hyperscaler and public cloud customers, while neocloud, industrial, and enterprise customers paid the remaining $37 billion. Edge sales accounted for a mere $6.4 billion, with the company citing demand for Blackwell-based workstation gear as a key driver. Looking ahead to Q2, Nvidia forecast revenue will hit $91 billion plus or minus two percent. That prediction assumed no datacenter sales in China. Nvidia has been trying for months to reignite its GPU business in the Middle Kingdom since Uncle Sam gave the company the green light to sell its aging H200 processors to Chinese customers for the first time ever back in December. Despite receiving approval from the Trump administration and receiving billions of dollars worth of orders, shipments remain stuck in Beijing's red tape. ®

Analysts home in on Nvidia's inference market share following an earnings win. Here's why

Artificial intelligence chipmaker Nvidia boosted its dividend, delivered a beat on earnings and revenue, and reported adjusted gross margins of 75% - but the stock still sank in extended trading after quarterly results released Wednesday. The "agentic" phase of the AI buildout has put the focus back on central processing units and more decentralized systems architecture, sending other chipmakers like Intel , Micron and AMD soaring in recent months while Nvidia performed a bit more modestly . NVDA AMD,INTC,MU 3M mountain Nvidia vs. Intel, AMD and Micron in the past three months Analysts wanted to know on Wednesday's earnings call whether Nvidia will keep its edge as the playing field for chip demand broadens out. "With investors keenly focused on your market share in inference, how do you see Vera Rubin ... impacting your share of the inference market?" CJ Muse, managing director at Cantor Fitzgerald, asked Nvidia CEO Jensen Huang. Vera Rubin is Nvidia's upcoming AI system , slated to rollout later this year. Huang responded not once, not twice but five or six times that the company's market share of inference - which refers to AI's ability to answer questions and complete tasks for users and companies - was growing. "Our share of inference is growing very quickly," he said. Nvidia is not the only kid on the block anymore when it comes to producing AI chips, which were synonymous with the graphics processing units that Nvidia specialized in. The newer "agentic" phase of AI - which refers to semi-autonomous bots that can be sent off to do tasks on their own, rather than more directly managed - requires a more distributed processing design that requires a greater number of more traditional CPUs than GPUs. While Nvidia is still undoubtedly the dominant player, the new infrastructure format is amping analyst interest in competition. It's not clear yet if a specific kind of architecture is going to dominate the market or if chipmakers will increasingly specialize. Rival custom silicon products like Alphabet's tensor processing units and Amazon's Trainium chips, along with products from AMD and newcomer Cerebras , could potentially disrupt this market. However, Huang said that production bandwidth for Nvidia's forthcoming Vera Rubin chip platform could be at capacity even before it ships later this year. "My sense is that we'll be supply constrained throughout the entire life of Vera Rubin," he said. Analysts are not seeing CPU or memory demand tapering any time soon. "Server CPU demand is strongly inflecting up, driving demand ... for both conventional server as well as AI servers (CPU head nodes). Agentic AI is a key driver," Nicolas Gaudois, analyst with UBS, wrote on Wednesday.

Vera Arrives: NVIDIA's First CPU Built for Agents Lands at Top AI Labs

Ian Buck hand-delivers the first NVIDIA Vera CPU systems to Anthropic, OpenAI, Oracle Cloud Infrastructure and SpaceXAI -- marking the moment agentic CPUs move from announcement to production. Agentic AI has always called for a different kind of CPU. NVIDIA CEO and founder Jensen Huang introduced the answer -- the standalone Vera CPU -- at GTC San Jose in March as NVIDIA's next multi-billion dollar business. On Friday, that CPU went from NVIDIA's labs into customer hands. The first NVIDIA Vera CPUs arrived at three of the world's leading AI labs on Friday -- Anthropic in San Francisco, OpenAI in Mission Bay, SpaceXAI in Palo Alto -- followed by a delivery to Oracle Cloud Infrastructure in Santa Clara on Monday. NVIDIA Vice President of Hyperscale and High-Performance Computing Ian Buck hand-delivered them. "Agentic AI is creating a new CPU moment in the AI factory -- as models move from answering to acting, Vera is purpose-built to keep that work moving at scale," Buck said. The big idea: imagine you could work 10x faster. Could your computer keep up? Agentic AI puts more demand than ever on the infrastructure we use to do all kinds of work -- from building slides to compiling and testing software, analyzing data, searching files or even running simulations. AI agents don't run on GPUs alone. Every agentic sandbox, every tool call, every orchestration layer, every long-context retrieval operation -- that's CPU work. Vera is a new class of CPU designed with that reality as its starting point. This gauntlet of concurrent, real-time tasks puts pressure on CPUs in ways traditional core-density focused designs were never built to prioritize. Vera packs 88 custom NVIDIA-designed Olympus cores, 1.2 TB/s of memory bandwidth and 50% faster per-core performance. Under constant load, work completes more quickly -- increasing the efficiency of the entire AI factory and helping users get their work done with faster responses. Vera Heads to San Francisco and Anthropic The first delivery landed at Anthropic's sleek SoMa offices in San Francisco. James Bradbury, Anthropic's head of compute, took the handoff from their conference room near the Bay.

'Every single frontier model company will jump on Vera Rubin from the get go': Nvidia CEO insists the future looks bright

It's hardly surprising that there was almost no mention of the gaming side of the business in Nvidia's latest quarterly earnings call. The company has been riding high on all things AI for several years now, to the tune of a $5.41 trillion market capitalisation as I write this. Unsurprisingly, CEO Jensen Huang is excited about what role the company's hardware will play in the industry's future. "Vera Rubin is going to be even more successful than Grace Blackwell at this point," Huang said in response to an investor question. "Every single frontier model company will jump on Vera Rubin from the get go -- and that was not true before on Blackwell. Vera Rubin is off to a tremendous start, and it will surely be more successful than even Grace Blackwell." Huang had earlier said that Nvidia is enjoying a "growing share in inference." The CEO credits an increase in frontier AI model companies, each one looking for appropriate hardware to power their projects. Huang specifically highlighted Nvidia's partnership with Anthropic, saying, "They are expanding incredibly fast. We have partnered with them to secure computing capacity across Azure, AWS, CoreWeave [etc]." For those who need the refresher, Vera Rubin is Nvidia's six-trillion transistor AI 'superchip'. It enjoys 100 times as much compute as the Nvidia GB10 Grace Blackwell Superchip that powers the Nvidia DGX Spark AI box. To be clear, though, 'Blackwell' also refers to the microarchitecture on which the RTX 50 series of GPUs are built. Vera Rubin will obviously be a far more specialised bit of kit, capable of rustling up 100 petaflops of raw compute and enjoying 60 times as many transistors as an RTX 5090 graphics card. Anyway, long story short, rapidly expanding AI partners mean that Nvidia, too, is on a roll. Huang said on the same call, "Demand for AI infrastructure continues to expand at an unprecedented pace. The build out of AI factories is accelerating. The value of NVIDIA AI infrastructure is rising." Such enthusiasm is within the company's own interest, though. For one thing, the first hint that the company's continued growth is about to flatten could result in the share price dipping. If you're looking for a more nuanced view from someone outside of Nvidia, our Jacob casts doubt on the company line and chats to a number of experts about the future of the AI industry. The mere fact that even the experts offer diverging visions suggests that AI's future is far from certain.

NVIDIA's 'Vera' CPUs could outperform and outsell competing x86 offerings from Intel and AMD

NVIDIA introduced its Vera CPU back in March at GTC 2026, billing it as the world's first processor purpose-built for agentic AI workloads. The chip is an ARM-based, 88-core design that can be sold standalone or paired with the Rubin GPU inside NVIDIA's NVL72 systems, and the company has already hand-delivered its first Vera units to partners including Anthropic, OpenAI, SpaceX, and Oracle. Now, the revenue projections attached to Vera are turning heads in a big way. During NVIDIA's recent earnings call, CFO Colette Kress stated that the company expects its Grace and Vera CPU lineup to bring in around $20 billion in revenue this fiscal year alone. CEO Jensen Huang framed the broader opportunity as a brand-new $200 billion total addressable market (TAM), one that NVIDIA says it has never previously addressed. To put that $20 billion figure in perspective, Intel's Data Center and AI division pulled in $16.8 billion across all of last year, while AMD's data center segment brought in $16.63 billion in 2025. However, one can argue that the AI boom will lead to higher revenue for these companies as we move through 2026 and towards 2027. For comparison, in Q1 2026, Intel's data center segment pulled in $5.1 billion, while AMD generated $5.8 billion. These figures comprise both Intel Xeon and AMD EPYC CPUs, along with accompanying products such as AMD's Instinct GPUs. If NVIDIA hits its target, it would effectively outsell both rivals in the CPU space within a single fiscal year, which is a remarkable statement for a company that has only just entered the market. Beyond sales, analysts at GF Securities are expecting NVIDIA to make a strong performance case for Vera at Computex 2026 on June 1. Their analysis suggests NVIDIA will claim Vera delivers around 1.5 times faster inference performance over competing x86 chips from Intel and AMD, alongside double the overall throughput and four times the rack density. Intel is expected to respond at the same show by spotlighting its entry-level Wildcat Lake CPUs, while AMD continues pushing its EPYC roadmap forward. As for when Vera reaches scale, GF Securities projects shipments of roughly 1.2 million units in FY2027, ramping sharply to about 4.2 million units in FY2028. Of course, this growth will be largely driven by the shift toward CPU-heavy agentic AI infrastructure. It will be interesting to see how AMD and Intel plan to compete with NVIDIA going forward. After dominating the GPU space, NVIDIA looks poised to eat their lunch money in the enterprise CPU market as well.

Jensen Huang Eyes CPU Boom as Agentic A.I. Reshapes Chip Market

After dominating GPUs, Nvidia is scaling its CPU business with new chips like Vera to capture growing demand from agentic A.I. systems. Nvidia's astronomical rise to the world's most valuable public company. was propelled by its graphics processing units (GPUs). Now, as A.I. enters its agentic phase, CEO Jensen Huang is betting future growth on central processing units (CPUs), long a secondary focus for the company. Sign Up For Our Daily Newsletter Sign Up Thank you for signing up! By clicking submit, you agree to our <a href="http://observermedia.com/terms">terms of service</a> and acknowledge we may use your information to send you emails, product samples, and promotions on this website and other properties. You can opt out anytime. See all of our newsletters "The world has 1 billion human users. My sense is that the world is going to have billions of agents. We're going to need a lot more CPUs," Huang told analysts on Nvidia's latest quarterly earnings call yesterday (May 20). GPUs are great at the parallel processing required for A.I. workloads, while CPUs excel at processing tokens quickly -- a key requirement for agentic systems. Nvidia has already generated $20 billion in CPU revenue this year, according to CFO Colette Kress, who told analysts these results are "setting us up to become the world's leading CPU supplier." A key piece of that push is Vera, Nvidia's newest CPU unveiled in March. Huang described it as a "major new growth driver," adding that the chips open up a "market we have never addressed before." He estimates the CPU market opportunity to be worth $200 billion. For now, Nvidia's bottom line continues to be driven by GPUs. For the February-April quarter, the company reported $81.6 billion in revenue, up 85 percent year-over-year, and a 211 percent jump in net income to $58.3 billion, beating Wall Street expectations. Data center revenue, which includes A.I. chips, reached $75 billion, up 92 percent from a year earlier. Huang has made a career out of anticipating industry shifts. When he founded Nvidia in 1993, it was a gaming-focused company. He later pivoted toward A.I. as GPUs proved well-suited for the technology, positioning Nvidia at the center of the current boom. More recently, the company has expanded into emerging areas like "physical A.I.", including autonomous vehicles and robotics. The CPU market, however, presents a different challenge. Rivals have already been gaining ground as demand for these chips rises alongside agentic A.I. workloads. Intel's stock has climbed 193 percent since January, while AMD is up 97 percent over the same period. Both companies are leaning into the shift. Intel CEO Lip-Bu Tan called CPUs the "indispensable foundation of the A.I. era," adding that the trend is "not just our wishful thinking, it's what we hear from customers." The company reported $13.6 billion in quarterly revenue last month, up 7 percent year-over-year. AMD is seeing similar momentum, with revenue rising 38 percent to $10.3 billion; CEO Lisa Su told analysts she expects CPU revenue to grow 70 percent this quarter. Still, analysts see opportunity for Nvidia. Its Vera chips are "adding a dimension to growth with the CPU market," said Raymond James analyst Simon Leopold. "While we are not certain that the ratio of CPUs to accelerators will go to 1:1, we accept the CPU growth is faster and the ratio is narrowing, and this is favorable for Nvidia."

NVIDIA hand-delivers the first Vera CPU systems to Anthropic, OpenAI, SpaceXAI, and more

NVIDIA has begun hand-delivering its first custom CPU designed for Agentic AI. The Vera CPU featured 88 custom Olympus cores, an impressive 1.2 TB/s of memory bandwidth, and 50% faster per-core performance under full load. And with that, NVIDIA Vice President of Hyperscale and High-Performance Computing, Ian Buck, personally hand-delivered the first Vera CPUs to the biggest AI companies in California: Anthropic, OpenAI, SpaceXAI, and Oracle Cloud Infrastructure. Although NVIDIA is synonymous with generative AI thanks to its GPUs, Agentic AI still requires a powerful CPU to handle tasks such as long-context retrieval. "Vera is a new class of CPU designed with that reality as its starting point," NVIDIA explains. "This gauntlet of concurrent, real-time tasks puts pressure on CPUs in ways traditional core-density focused designs were never built to prioritize." As the successor to Grace, Vera underscores the growing importance of CPUs in AI, and the Vera CPU is the first data center CPU to use LPDDR5X memory, delivering 1.2 TB/s of memory bandwidth. And with that, even though it will ship as part of the Vera Rubin platform that combines CPU and GPU technology, it's also set to be made available as a standalone CPU solution. NVIDIA's hand-delivery of the first Vera CPUs signals that Vera and Rubin are entering their respective full production phases for a global rollout in the coming months. "Agentic AI is creating a new CPU moment in the AI factory - as models move from answering to acting, Vera is purpose-built to keep that work moving at scale," NVIDIA's Ian Buck said. "When AI models are posed a question, the answer, often, isn't already prepped and ready to go. The models actually have to generate some Python code to arrive at the correct answer. A task at which the Vera CPU excels. That's why we are seeing the demand for CPUs skyrocket."

NVIDIA's Vera CPUs Set to Outrun Intel and AMD x86 Chips by 1.5x at Computex 2026, Analysts Predict

With the Computex 2026 trade show in Taiwan coming up, analysts from GF Securities are confident that NVIDIA will address the latest trend in the AI computing industry at the event. Following Intel's first-quarter earnings, the industry's shift towards CPUs for agentic AI computing has generated speculation that the role that GPUs play within an AI cluster could drop. However, GF Securities believes that NVIDIA will showcase the advantages of its Vera CPUs at Computex and claim that they offer as much as 1.5 times the computing performance over x86 rivals from AMD and Intel. At the center of the analysts' coverage is the role of Arm-based CPUs in the new rush for CPUs for agentic AI computing. At the upcoming Computex show in Taiwan, they expect NVIDIA to focus on its Vera Rubin CPUs and Intel to focus on the entry-level Wildcat CPUs. Both these announcements will cater to the competitive challenges that the two semiconductor companies are facing. For NVIDIA, the announcements are expected to demonstrate the firm's strengths in the evolving AI market that is now focused on using CPUs for agentic AI computing. As per the analysts, at Computex, NVIDIA will showcase the advantages of its Vera CPUs. They add that the potential impact of inferencing workloads on the CPU-to-GPU ratio could also drive Vera adoption higher. Specifically, the analysts expect NVIDIA to reveal that the Vera CPUs are capable of delivering 1.5x faster performance, 2x overall performance and 4x higher density per rack compared to the rival x86 chips from Intel and NVIDIA. They add that the firm could ship 1.2 million CPU units in fiscal year 2027 and 4.2 million units in fiscal year 2028. As for Intel, GF Securities expects the firm to focus on its Wildcat Lake chips. They expect these chips to be launched as a competitive option to Apple's MacBook Neo laptops, which are the entry-level computers in the Cupertino, California-based consumer electronics firm's notebook portfolio. Overall, the analysts also expect the growth in agentic AI to significantly increase the total addressable market for server CPUs. With agentic AI expected to account for 30% of overall inference computing in the future, the CPU TAM can touch $211 billion by 2030, they explain. In terms of units, the TAM growth should see the CPU demand grow from 3.7 million units in 2026 to 16.3 million units in 2028.

NVIDIA's Next Gold Mine? Jensen Huang Sees 'Brand New' $200 Billion Opportunity In Vera CPUs - Intel (NAS

Jensen Huang, CEO of NVIDIA Corporation (NASDAQ:NVDA), has unveiled a new $200 billion market for the company, courtesy of its latest product, the Vera CPU. "Vera opens a brand new $200 billion TAM for Nvidia," Huang said. The CEO expressed that Vera has tapped into a "never addressed before" market. The company is collaborating with every major hyperscaler and system maker to deploy it. He stressed that NVIDIA is at the core of the shift towards agentic AI and robotic physical AI. Huang explained that the company's Vera CPU platform will be deployed in multiple configurations, including alongside Rubin GPUs, as a standalone CPU, and integrated with ConnectX-9 networking for storage and security workloads. He added that millions of Rubin systems paired with Vera CPUs are expected to ship, with additional use cases focused on confidential computing, compute isolation, and infrastructure software. While AI models use GPUs for "thinking," agents primarily run on CPUs to perform their assigned tasks. Huang anticipates that these agents will operate their own form of CPU-driven PCs. He said NVIDIA has already sold $20 billion worth of standalone Vera CPUs this year, calling it the beginning of a new era. With billions of agents predicted to use tools similar to PCs, Huang believes the world will "need a lot more CPUs," presenting a massive opportunity for NVIDIA. AI Inference Fuels CPU Demand This news comes on the heels of NVIDIA's announcement that its first Vera CPU systems were delivered to leading AI companies, including SpaceX, Anthropic, and OpenAI. This move is part of NVIDIA's broader push into next-generation infrastructure designed for "agentic AI" workloads. According to Benzinga Edge Stock Rankings, Nvidia has a growth score of 98.25% and a momentum rating of 85.05%. Benzinga's screener allows you to compare NVDA's performance with its peers. NVDA Price Action: On a year-to-date basis, Nvidia stock climbed 18.33%, as per Benzinga Pro. On Wednesday, it closed 1.30% higher at $223.47. Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors. Image via Shutterstock Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.

NVIDIA Hand-Delivers First Vera CPUs to Anthropic, OpenAI, SpaceX and Oracle as Grace's Successor Powers the Agentic AI Era

NVIDIA has started shipping the very first Vera CPUs to major AI firms, marking the official beginning of production to accelerate Agentic AI forward. The Vera CPU is the next chapter for NVIDIA's Agentic AI ecosystem, setting the stage for its next multi-billion dollar frontier. Today, NVIDIA's Vice President of Hyperscale and High-Performance Computing, Ian Buck, hand-delivered the first Vera CPUs to four major AI firms who are accelerating Agentic AI forward, these include OpenAI, SpaceX, Anthropic, and Oracle Cloud. This marks the first time that the Vera CPU left NVIDIA's labs and landed in the hands of its first customers. Ian shares his journey in delivering the Vera CPUs. The first Vera CPUs landed at Anthropic's SoMa offices in San Francisco. Next stop, OpenAI's Mission Bay HQ, where the second rack featuring Vera CPUs was delivered. And lastly, SpaceXAI, where Elon Musk himself was handed Vera at the Palo Alto office. All three Vera racks were delivered on Friday. On Monday, Ian made a visit to Oracle's AI Customer Excellence Center, where he delivered the last Vera CPU rack. These deliveries are just a tiny portion of what NVIDIA will be shipping in the coming quarters. Agentic AI firms are showing massive interest in CPUs these days, and Vera, being built for it, is driving crazy traction. Just last week, we reported that Vera CPUs were going to be adopted by CoreWeave, Meta, and Alibaba as early buyers. Oracle was also mentioned, and as you can see, they have already received the first chips. According to NVIDIA, the Vera CPU offers extremely high single-threaded core performance, incredibly high data output, and extreme levels of energy efficiency. Vera is the world's first and only data center CPU to utilize LPDDR5 memory and offers unrivaled performance per watt. NVIDIA is not just integrating Vera CPUs into its Vera Rubin platform; these will also be shipped standalone, & the company expects this to open another multi-billion-dollar business front for it. For the Vera CPU, NVIDIA has designed its next-gen custom Arm architecture codenamed Olympus, and the chip packs 88 cores, 176 threads (with NVIDIA Spatial Multi-Threading), 1.8 TB/s NVLink-C2C coherent memory interconnect, 1.5 TB of system memory (3x Grace), 1.2 TB/s of memory bandwidth with SOCAMM LPDDR5X, and Rack-scale confidential compute. These combine to offer 2x data processing, compression & CI/CD performance versus Grace. With Vera, we will also see demand for LPDDR5X DRAM swell as the platform features support for up to 1.5 TB of memory, and given the amount of Vera CPUs that will be required to meet compute demands, the supply chain is going to see increased constraints. NVIDIA's Vera Rubin platform is set for launch in the second half of this week, with mass production of the first racks commencing real soon. Vera will be used in both standalone LPX servers and as a host processor in the Vera Rubin NVL72 racks. With Vera now entering the full production phase, and Rubin all set for launch in the coming months, the extreme co-design platform that NVIDIA has come up with is going to propel agentic AI workflows to the next level.