Operator:

Good afternoon. My name is Rob and I'll be your conference operator today. At this time, I would like to welcome everyone to the NVIDIA's Fourth Quarter Earnings Call. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there will be a question-and-answer session. [Operator Instructions] Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. Q4 was another record quarter. Revenue of $22.1 billion was up 22% sequentially and up to 265% year-on-year and well above our outlook of $20 billion. For fiscal 2024, revenue was $60.9 billion and up 126% from the prior year. Starting with data center. Data center revenue for the fiscal 2024 year was $47.5 billion, more than tripling from the prior year. The world has reached the tipping point of new computing era. The $1 trillion installed base of data center infrastructure is rapidly transitioning from general purpose to accelerated computing. As Moore's Law slows while computing demand continues to skyrocket, companies may accelerate every workload possible to drive future improvement in performance, TCO and energy efficiency. At the same time, companies have started to build the next generation of modern data centers, what we refer to as AI factories, purpose built to refine raw data and produce valuable intelligence in the era of generative AI. In the fourth quarter, data center revenue of $18.4 billion was a record, up 27% sequentially and up 409% year-over-year, driven by the NVIDIA Hopper GPU computing platform along with InfiniBand end-to-end networking. Compute revenue grew more than 5x and networking revenue tripled from last year. We are delighted that supply of Hopper architecture products is improving. Demand for Hopper remains very strong. We expect our next-generation products to be supply constrained as demand far exceeds supply. Fourth quarter data center growth was driven by both training and inference of generative AI and large language models across a broad set of industries, use cases and regions. The versatility and leading performance of our data center platform enables a high return on investment for many use cases, including AI training and inference, data processing and a broad range of CUDA accelerated workloads. We estimate in the past year approximately 40% of data center revenue was for AI inference. Building and deploying AI solutions has reached virtually every industry. Many companies across industries are training and operating their AI models and services at scale, enterprises across NVIDIA AI infrastructure through cloud providers, including hyperscales, GPU specialized and private clouds or on-premise. NVIDIA's computing stack extends seamlessly across cloud and on-premise environments, allowing customers to deploy with a multi-cloud or hybrid-cloud strategy. In the fourth quarter, large cloud providers represented more than half of our data center revenue, supporting both internal workloads and external public cloud customers. Microsoft recently noted that more than 50,000 organizations use GitHub Copilot business to supercharge the productivity of their developers, contributing to GitHub revenue growth accelerating to 40% year-over-year. And Copilot for Microsoft 365 adoption grew faster in its first two months than the two previous major Microsoft 365 enterprise suite releases did. Consumer internet companies have been early adopters of AI and represent one of our largest customer categories. Companies from search to e-commerce, social media, news and video services and entertainment are using AI for deep learning-based recommendation systems. These AI investments are generating a strong return by improving customer engagement, ad conversation and click-throughs rates. Meta in its latest quarter cited more accurate predictions and improved advertiser performance as contributing to the significant acceleration in its revenue. In addition, consumer internet companies are investing in generative AI to support content creators, advertisers and customers through automation tools for content and ad creation, online product descriptions and AI shopping assistance. Enterprise software companies are applying generative AI to help customers realize productivity gains. Early customers we've partnered with for both training and inference of generative AI are already seeing notable commercial success. ServiceNow's generative AI products in their latest quarter drove their largest ever net new annual contract value contribution of any new product family release. We are working with many other leading AI and enterprise software platforms as well, including Adobe, Databricks, Getty Images, SAP and Snowflake. The field of foundation of large-language models is thriving. Anthropic, Google, Inflection, Microsoft, OpenAI and xAI are leading with continued amazing breakthrough in generative AI. Exciting companies like Adept, AI21, Character.ai, Cohere, Mistral, Perplexity and Runway are building platforms to serve enterprises and creators. New startups are creating LLMs to serve the specific languages, cultures and customs of the world many regions. And others are creating foundation models to address entirely different industries like Recursion Pharmaceuticals and Generate:Biomedicines for biology. These companies are driving demand for NVIDIA AI infrastructure through hyperscale or GPU specialized cloud providers. Just this morning, we announced that we've collaborated with Google to optimize its state-of-the art new Gemma language models to accelerate their inference performance on NVIDIA GPUs in the cloud data center and PC. One of the most notable trends over the past year is the significant adoption of AI by enterprises across the industry verticals such as automotive, healthcare and financial services. NVIDIA offers multiple application frameworks to help companies adopt AI in vertical domains such as autonomous driving, drug discovery, low latency machine learning for fraud detection or robotics, leveraging our full stack accelerated computing platform. We estimate the data center revenue contribution of the automotive vertical through the cloud or on-prem exceeded $1 billion last year. NVIDIA DRIVE infrastructure solutions includes systems and software for the development of autonomous driving, including data ingestion, creation, labeling and AI training, plus validation through simulation. Almost 80 vehicle manufacturers across global OEMs, new energy vehicles, trucking, robotaxi and Tier 1 suppliers are using NVIDIA's AI infrastructure to train LLMs and other AI models for automated driving and AI cockpit applications. And in fact, nearly every automotive company working on AI is working with NVIDIA. As AV algorithms move to video transformers and more cars are equipped with cameras, we expect NVIDIA's automotive data center processing demand to grow significantly. In healthcare, digital biology and generative AI are helping to reinvent drug discovery, surgery, medical imaging and wearable devices. We have built deep domain expertise in healthcare over the past decade, creating the NVIDIA Clara healthcare platform and NVIDIA BioNeMo, a generative AI service to develop, customize and deploy AI foundation models for computer-aided drug discovery. BioNeMo features a growing collection of pre-trained Biomolecular AI models that can be applied to the end-to-end drug discovery processes. We announced Recursion is making available for their proprietary AI model through BioNeMo for the drug discovery ecosystem. In financial services, customers are using AI for a growing set of use cases from trading and risk management to customer service and fraud detection. For example, American Express improved fraud detection accuracy by 6% using NVIDIA AI. Shifting to our data center revenue by geography. Growth was strong across all regions, except for China where our data center revenue declined significantly following the U.S. government export control regulations imposed in October. Although we have not received licenses from the U.S. government to ship restricted products to China, we have started shipping alternatives that don't require a license for the China market. China represented a mid-single digit percentage of our data center revenue in Q4. And we expect it to stay in a similar range in the first-quarter. In regions outside of the U.S. and China, sovereign AI has become an additional demand driver. Countries around the world are investing in AI infrastructure to support the building of large-language models in their own language, on domestic data and in support of their local research and enterprise ecosystems. From a product perspective, the vast majority of revenue was driven by our Hopper architecture along with InfiniBand networking. Together, they have emerged as the de-facto standard for accelerated computing and AI infrastructure. We are on track to ramp H200 with initial shipments in the second quarter. Demand is strong as H200 nearly doubles the inference performance of H100. Networking exceeded a $13 billion annualized revenue run rate. Our end-to-end networking solutions define modern AI data centers. Our Quantum InfiniBand solutions grew more than 5x year on year. NVIDIA Quantum InfiniBand is the standard for the highest performance AI-dedicated infrastructures. We are now entering the ethernet networking space with the launch of our new Spectrum-X end-to-end offering designed for an AI-optimized networking for the data center. Spectrum-X introduces new technologies over ethernet, that are purpose built for AI. Technologies incorporated in our Spectrum switch, BlueField DPU and software stack deliver 1.6x higher networking performance for AI processing compared with traditional ethernet. Leading OEMs, including Dell, HPE, Lenovo and Super Micro, with their global sales channels, are partnering with us to expand our AI solution to enterprises worldwide. We are on track to ship Spectrum-X this quarter. We also made great progress with our software and services offerings, which reached an annualized revenue run rate of $1 billion in Q4. We announced that NVIDIA DGX Cloud will expand its list of partners to include Amazon's AWS, joining Microsoft Azure, Google Cloud and Oracle Cloud. DGX Cloud is used for NVIDIA's own AI R&D and custom model development as well as NVIDIA developers. It brings the CUDA ecosystem to NVIDIA CSP partners. Okay, moving to gaming. Gaming revenue was $2.87 billion, was flat sequentially and up 56% year on year, better than our outlook on solid consumer demand for NVIDIA GeForce RTX GPUs during the holidays. Fiscal year revenue of $10.45 billion was up 15%. At CES, we announced our GeForce RTX 40 Super Series family of GPUs. Starting at $599, they deliver incredible gaming performance and generative AI capabilities. Sales are off to a great start. NVIDIA AI Tensor cores and the GPUs deliver up to 836 AI tops, perfect for powering AI for gaming, creating an everyday productivity. The rich software stack we offer with our RTX GPUs further accelerates AI. With our DLSS technologies, seven out of eight pixels can be AI generated, resulting up to 4x faster ray tracing and better image quality. And with the Tensor RT LLM for Windows, our open-source library that accelerates inference performance for the latest large-language models generative AI can run up to 5X faster on RTX AI PCs. At CES, we also announced a wave of new RTX 40 Series AI laptops from every major OEMs. These bring high-performance gaming and AI capabilities to a wide range of form factors, including 14 inch and thin and light laptops. With up to 686 tops of AI performance, these next-generation AI PCs increase generative AI performance by up to 60x, making them the best-performing AI PC platforms. At CES, we announced NVIDIA Avatar Cloud Engine microservices, which allowed developers to integrate state-of-the-art generative AI models into digital avatars. ACE won several Best of CES 2024 awards. NVIDIA has an end-to-end platform for building and deploying generative AI applications for RTX PCs and workstations. This includes libraries, SDKs, tools and services developers can incorporate into their generative AI workloads. NVIDIA is fueling the next wave of generative AI applications coming to the PC. With over 100 million RTX PCs in the installed-base and over 500 AI-enabled PC applications and games, we are on our way. Moving to Pro Visualization. Revenue of $463 million was up 11% sequentially and up 105% year on year. Fiscal year revenue of $1.55 billion was up 1%. Sequential growth in the quarter was driven by a rich mix of RTX Ada architecture GPUs continuing to ramp. Enterprises are refreshing their workstations to support generative AI-related workloads, such as data preparation, LLM fine-tuning and retrieval augmented generation. These key industrial verticals driving demand include manufacturing, automotive and robotics. The automotive industry has also been an early adopter of NVIDIA Omniverse as it seeks to digitize work flows from design to build, simulate, operate and experience their factories and cars. At CES, we announced that creative partners and developers including Brickland, WPP and ZeroLight are building Omniverse-powered car configurators. Leading automakers like LOTUS are adopting the technology to bring new levels of personalization, realism and interactivity to the car buying experience. Moving to Automotive. Revenue was $281 million, up 8% sequentially and down 4% year on year. Fiscal year revenue of $1.09 billion was up 21%, crossing the $1 billion mark for the first time on continued adoption of the NVIDIA DRIVE platform by automakers. NVIDIA DRIVE Orin is the AI car computer of choice for software-defined AV fleets. Its successor, NVIDIA DRIVE Thor, designed for vision transformers often -- offers more AI performance and integrates a wide range of intelligent capabilities into a single AI compute platform, including autonomous driving and parking, driver and passenger monitoring and AI cockpit functionality and will be available next year. There were several automotive customer announcements this quarter, Li Auto, Great Wall Motor, ZEEKR, the premium EV subsidiary of Geely and Jeremy Xiaomi EV all announced new vehicles built on NVIDIA. Moving to the rest of the P&L. GAAP gross margins expanded sequentially to 76% and non-GAAP gross margins to 76.7% on strong data center growth and mix. Our gross margins in Q4 benefited from favorable component costs. Sequentially, GAAP operating expenses were up 6% and non-GAAP operating expenses were up 9%, primarily reflecting higher compute and infrastructure investments and employee growth. In Q4, we returned $2.8 billion to shareholders in the form of share repurchases and cash dividends. During fiscal year '24, we utilized cash of $9.9 billion towards shareholder returns, including $9.5 billion in share repurchases. Let me turn to the outlook for the first quarter. Total revenue is expected to be $24 billion, plus or minus 2%. We expect sequential growth in data center and proviz, partially offset by seasonal decline in gaming. GAAP and non-GAAP gross margins are expected to be 76.3% and 77% respectively, plus or minus 50 basis-points. Similar to Q4, Q1 gross margins are benefiting from favorable component costs. Beyond Q1, for the remainder of the year, we expect gross margins to return to the mid-70s percent range. GAAP and non-GAAP operating expenses are expected to be approximately $3.5 billion and $2.5 billion respectively. Fiscal year 2025 GAAP and non-GAAP operating expenses are expected to grow in the mid-30% range as we continue to invest in the large opportunities ahead of us. GAAP and non-GAAP other income and expenses are expected to be an income of approximately $250 million, excluding gains and losses from non-affiliated investments. GAAP and non-GAAP tax rates are expected to be 17%, plus or minus 1% excluding any discrete items. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight some upcoming events for the financial community. We will attend the Morgan Stanley Technology and Media and Telecom Conference in San Francisco on March 4 and the TD Cowen's 44th Annual Healthcare Conference in Boston on March 5. And of course, please join us for our Annual DTC conference starting Monday March 18 in San Jose, California, to be held in-person for the first time in five years. DTC will kick off with Jen-Hsun's keynote and we will host a Q&A session for financial analysts the next day, March 19. At this time, we will now open the call for questions. Operator, would you please poll for questions?

Toshiya Hari:

Hi. Thank you so much for taking the question and congratulations on the really strong results. My question is for Jen-Hsun on the data center business. Clearly, you're doing extremely well in the business. I'm curious how your expectations for calendar '24 and '25 have evolved over the past 90 days. And as you answer the question, I was hoping you can touch on some of the newer buckets within data center, things like software. Sovereign AI, I think you've been pretty vocal about how to think about that medium-to-long term. And recently, there was an article about NVIDIA potentially participating in the ASIC market. Is there any credence to that, and if so, how should we think about you guys playing in that market over the next several years? Thank you.

Toshiya Hari:

I guess your expectations for data center, how they've evolved. Thank you.

Operator:

Your next question comes from the line of Joe Moore from Morgan Stanley. Your line is open.

Jensen Huang:

I'll go backwards. The estimate is probably understated. And -- but we estimated it. And let me tell you why. Whenever -- a year ago, the recommender systems that people are -- when you run the internet, the news, the videos, the music, the products that are being recommended to you because as you know, the internet has trillions -- I don't know how many trillions, but trillions of things out there and your phone is 3-inches square. And so the ability for them to fit all of that information down to something, such a small real estate, is through a system, an amazing system called recommender systems. These recommender systems used to be all based on CPU approaches. But the recent migration to deep learning and now generative AI has really put these recommender systems now directly into the path of GPU acceleration. It needs GPU acceleration for the embeddings. It needs GPU acceleration for the nearest neighbor search. It needs GPU acceleration for the re-ranking and it needs GPU acceleration to generate the augmented information for you. So GPUs are in every single step of a recommender system now. And as you know, recommender system is the single largest software engine on the planet. Almost every major company in the world has to run these large recommender systems. Whenever you use ChatGPT, it's being inferenced. Whenever you hear about Midjourney and just the number of things that they're generating for consumers, when you when you see Getty, the work that we do with Getty and Firefly from Adobe. These are all generative models. The list goes on. And none of these, as I mentioned, existed a year ago, 100% new.

Stacy Rasgon:

Hi, guys. Thanks for taking my question. I wanted Colette -- I wanted to touch on your comment that you expected the next generation of products -- I assume that meant Blackwell, to be supply constrained. Could you dig into that a little bit, what is the driver of that? Why does that get constrained as Hopper is easing up? And how long do you expect that to be constrained, like do you expect the next generation to be constrained like all the way through calendar '25, like when do those start to ease?

Operator:

Your next question comes from the line of Matt Ramsay from TD Cowen. Your line is open.

Colette Kress:

Let me first start with your question, thanks, about how we are working with our customers as they look into how they are building out their GPU instances and our allocation process. The folks that we work with, our customers that we work with, have been partners with us for many years as we have been assisting them both in what they set up in the cloud, as well as what they are setting up internally. Many of these providers have multiple products going at one time to serve so many different needs across their end customers but also what they need internally. So they are working in advance, of course, thinking about those new clusters that they will need. And our discussions with them continue not only on our Hopper architecture, but helping them understand the next wave and getting their interest and getting their outlook for the demand that they want. So it's always a moving process in terms of what they will purchase, what is still being built and what is in use for our end customers. But the relationships that we've built and their understanding of the sophistication of the build has really helped us with that allocation and both helped us with our communications with them.

Operator:

Your next question comes from the line of Timothy Arcuri from UBS. Your line is open.

Colette Kress:

Yeah. So let me, highlight on those three different areas of how we look at our suppliers. You're correct. Our inventory on hand given our allocation that we're on, we're trying to, as things come into inventory, immediately work to ship them to our customers. I think our customer appreciates our ability to meet the schedules that we've looked for. The second piece of it is our purchase commitments. Our purchase commitments have many different components into it, components that we need for manufacturing. But also, often we are procuring capacity that we may need. The length of that need for capacity or the length for the components are all different. Some of them may be for the next two quarters, but some of them may be for multiple years. I can say the same regarding our prepaids. Our prepaids are pre-designed to make sure that we have the reserve capacity that we need at several of our manufacturing suppliers as we look forward. So wouldn't read into anything regarding approximately about the same numbers as we are increasing our supply. All of them just have different lengths as we have sometimes had to buy things in long-lead times or things that needed capacity to be built for us.

Ben Reitzes:

Yeah. Thanks. Congratulations on the results. Colette, I wanted to talk about your comment regarding gross margins and that they should go back to the mid-70s. If you don't mind unpacking that. And also, is that due to the HBM content in the new products and what do you think are the drivers of that comment? Thanks so much.

Operator:

Your next question comes from the line of C.J. Muse from Cantor Fitzgerald. Your line is open.

Jensen Huang:

Hey, CJ. Thanks for the question. Yeah, that's the really cool part. If you look at the reason why we're able to improve performance so much, it's because we have two characteristics about our platform. One, is that it's accelerated. And two, it's programmable. It's not brittle. NVIDIA is the only architecture that has gone from the very, very beginning, literally the very beginning when CNN's and Alex Krizhevsky and Ilya Sutskever and Geoff Hinton first revealed AlexNet, all the way through to RNNs to LSTMs to every -- RLs to deep learning RLs to transformers to every single version. Every single version and every species that have come along, vision transformers, multi-modality transformers, every single -- and now time sequence stuff, and every single variation, every single species of AI that has come along, we've been able to support it, optimize our stack for it and deploy it into our installed base. This is really the great amazing part. On the one hand, we can invent new architectures and new technologies like our Tensor cores, like our transformer engine for Tensor cores, improved new numerical formats and structures of processing like we've done with the different generations of Tensor cores, meanwhile, supporting the installed base at the same time. And so, as a result, we take all of our new software algorithm invest -- inventions, all of the inventions, new inventions of models of the industry, and it runs on our installed base on the one hand. On the other hand, whenever we see something revolutionary we can -- like transformers, we can create something brand new like the Hopper transformer engine and implement it into future. And so we simultaneously have this ability to bring software to the installed base and keep making it better and better and better, so our customers installed base is enriched over time with our new software. On the other hand, for new technologies, create revolutionary capabilities. Don't be surprised if in our future generation, all of a sudden amazing breakthroughs in large-language models were made possible And those breakthroughs, some of which will be in software because they run CUDA, will be made available to the installed base. And so we carry everybody with us on the one hand. We make giant breakthroughs on the other hand.

Aaron Rakers:

Yeah. Thanks for taking the question. I wanted to ask about the China business. I know that in your prepared comments you said that you started shipping some alternative solutions into China. You also put it out that you expect that contribution to continue to be about a mid-single digit percent of your total data center business. So I guess the question is what is the extent of products that you're shipping today into the China market and why should we not expect that maybe other alternative solutions come to the market and expand your breadth to participate in that in that opportunity again? Thank you.

Operator:

Your next question comes from the line of Harsh Kumar from Piper Sandler. Your line is open.

Jensen Huang:

Let me take a step back and explain the fundamental reason why NVIDIA will be very successful in software. So first, as you know, accelerated computing really grew in the cloud. In the cloud, the cloud service providers have really large engineering teams and we work with them in a way that allows them to operate and manage their own business. And whenever there are any issues, we have large teams assigned to them. And their engineering teams are working directly with our engineering teams and we enhance, we fix, we maintain, we patch the complicated stack of software that's involved in accelerated computing. As you know, accelerated computing is very different than general-purpose computing. You're not starting from a program like C++. You compile it and things run on all your CPUs. The stacks of software necessary for every domain from data processing SQL versus -- SQL structure data versus all the images and text and PDF, which is unstructured, to classical machine-learning to computer vision to speech to large-language models, all --recommender systems. All of these things require different software stacks. That's the reason why NVIDIA has hundreds of libraries. If you don't have software, you can't open new markets. If you don't have software, you can't open and enable new applications. Software is fundamentally necessary for accelerated computing. This is the fundamental difference between accelerated computing and general-purpose computing that most people took a long time to understand. And now, people understand that the software is really key. And the way that we work with CSPs, that's really easy. We have large teams that are working with their large teams. However, now that generative AI is enabling every enterprise and every enterprise software company to embrace accelerated computing -- and when -- it is now essential to embrace accelerated computing because it is no longer possible, no longer likely anyhow to sustain improved throughput through just general-purpose computing. All of these enterprise software companies and enterprise companies don't have large engineering teams to be able to maintain and optimize their software stack to run across all of the world's clouds and private clouds and on-prem. So we are going to do the management, the optimization, the patching, the tuning, the installed-base optimization for all of their software stacks. And we containerize them into our stack. We call it NVIDIA AI Enterprise. And the way we go to market with it is that think of that NVIDIA AI Enterprise now as a run time like an operating system, it's an operating system for artificial intelligence. And we charge $4,500 per GPU per year. And my guess is that every enterprise in the world, every software enterprise company that are deploying software in all the clouds and private clouds and on-prem, will run on NVIDIA AI Enterprise, especially obviously for our GPUs. And so this is going to likely be a very significant business over time. We're off to a great start. And Colette mentioned that it's already at $1 billion run rate and we're really just getting started.

Jensen Huang:

The computer industry is making two simultaneous platform shifts at the same time. The trillion-dollar installed base of data centers is transitioning from general purpose to accelerated computing. Every data center will be accelerated so the world can keep up with the computing demand, with increasing throughput, while managing costs and energy. The incredible speed up of NVIDIA enabled -- that NVIDIA enabled, a whole new computing paradigm, generative AI, where software can learn, understand and generate any information from human language to the structure of biology and the 3D world. We are now at the beginning of a new industry where AI-dedicated data centers process massive raw data to refine it into digital intelligence. Like AC power generation plants of the last industrial revolution, NVIDIA AI supercomputers are essentially AI generation factories of this Industrial Revolution. Every company in every industry is fundamentally built on their proprietary business intelligence, and in the future, their proprietary generative AI. Generative AI has kicked off a whole new investment cycle to build the next trillion dollars of infrastructure of AI generation factories. We believe these two trends will drive a doubling of the world's data center infrastructure installed base in the next five years and will represent an annual market opportunity in the hundreds of billions. This new AI infrastructure will open up a whole new world of applications not possible today. We started the AI journey with the hyperscale cloud providers and consumer internet companies. And now, every industry is on board, from automotive to healthcare to financial services, to industrial to telecom, media and entertainment. NVIDIA's full stack computing platform with industry-specific applications frameworks and a huge developer and partner ecosystem, gives us the speed, scale and reach to help every company -- to help companies in every industry become an AI company. We have so much to share with you at next month's GTC in San Jose. So be sure to join us. We look forward to updating you on our progress next quarter.

Operator:

Good afternoon. My name is JL, and I will be your conference operator today. At this time, I would like to welcome everyone to NVIDIA's Third Quarter Earnings Call. All lines have been placed on mute to prevent any background noise. After the speakers’ remarks, there will be a question-and-answer session. [Operator Instructions] Simona Jankowski, you may now begin your conference.

Colette Kress:

Thanks, Simona. Q3 was another record quarter. Revenue of $18.1 billion was up 34% sequentially and up more than 200% year-on-year and well above our outlook for $16 billion. Starting with Data Center. The continued ramp of the NVIDIA HGX platform based on our Hopper Tensor Core GPU architecture, along with InfiniBand end-to-end networking, drove record revenue of $14.5 billion, up 41% sequentially and up 279% year-on-year. NVIDIA HGX with InfiniBand together are essentially the reference architecture for AI supercomputers and data center infrastructures. Some of the most exciting generative AI applications are built and run on NVIDIA, including Adobe Firefly, ChatGPT, Microsoft 365 Copilot, CoAssist, now assist with ServiceNow and Zoom AI Companion. Our Data Center compute revenue quadrupled from last year and networking revenue nearly tripled. Investments in infrastructure for training and inferencing large language models, deep learning, recommender systems and generative AI applications is fueling strong broad-based demand for NVIDIA accelerated computing. Inferencing is now a major workload for NVIDIA AI computing. Consumer Internet companies and enterprises drove exceptional sequential growth in Q3, comprising approximately half of our Data Center revenue and outpacing total growth. Companies like Meta are in full production with deep learning, recommender systems and also investing in generative AI to help advertisers optimize images and text. Most major consumer Internet companies are racing to ramp up generative AI deployment. The enterprise wave of AI adoption is now beginning. Enterprise software companies such as Adobe, Databricks, Snowflake and ServiceNow are adding AI copilots and the systems to their platforms. And broader enterprises are developing custom AI for vertical industry applications such as Tesla in autonomous driving. Cloud service providers drove roughly the other half of our Data Center revenue in the quarter. Demand was strong from all hyperscale CSPs, as well as from a broadening set of GPU-specialized CSPs globally that are rapidly growing to address the new market opportunities in AI. NVIDIA H100 Tensor Core GPU instances are now generally available in virtually every cloud with instances in high demand. We have significantly increased supply every quarter this year to meet strong demand and expect to continue to do so next year. We will also have a broader and faster product launch cadence to meet the growing and diverse set of AI opportunities. Towards the end of the quarter, the U.S. government announced a new set of export control regulations for China and other markets, including Vietnam and certain countries in the Middle East. These regulations require licenses for the export of a number of our products, including our Hopper and Ampere 100 and 800 series and several others. Our sales to China and other affected destinations derived from products that are now subject to licensing requirements have consistently contributed approximately 20% to 25% of Data Center revenue over the past few quarters. We expect that our sales to these destinations will decline significantly in the fourth quarter. So we believe will be more than offset by strong growth in other regions. The U.S. government designed the regulation to allow the U.S. industry to provide data center compute products to markets worldwide, including China. Continuing to compete worldwide as the regulations encourage, promotes U.S. technology leadership, spurs economic growth and supports U.S. jobs. For the highest performance levels, the government requires licenses. For lower performance levels, the government requires a streamlined prior notification process. And for products even lower performance levels, the government does not require any notice at all. Following the government's clear guidelines, we are working to expand our Data Center product portfolio to offer compliance solutions for each regulatory category, including products for which the U.S. government does not wish to have advance notice before each shipment. We are working with some customers in China and the Middle East to pursue licenses from the U.S. government. It is too early to know whether these will be granted for any significant amount of revenue. Many countries are awakening to the need to invest in sovereign AI infrastructure to support economic growth and industrial innovation. With investments in domestic compute capacity, nations can use their own data to train LLMs and support their local generative AI ecosystems. For example, we are working with India's government and largest tech companies including Infosys, Reliance and Tata to boost their sovereign AI infrastructure. And French private cloud provider, Scaleway, is building a regional AI cloud based on NVIDIA H100 InfiniBand and NVIDIA's AI Enterprise software to fuel advancement across France and Europe. National investment in compute capacity is a new economic imperative and serving the sovereign AI infrastructure market represents a multi-billion dollar opportunity over the next few years. From a product perspective, the vast majority of revenue in Q3 was driven by the NVIDIA HGX platform based on our Hopper GPU architecture with lower contribution from the prior generation Ampere GPU architecture. The new L40S GPU built for industry standard servers began to ship, supporting training and inference workloads across a variety of consumers. This was also the first revenue quarter of our GH200 Grace Hopper Superchip, which combines our ARM-based Grace CPU with a Hopper GPU. Grace and Grace Hopper are ramping into a new multi-billion dollar product line. Grace Hopper instances are now available at GPU specialized cloud providers, and coming soon to Oracle Cloud. Grace Hopper is also getting significant traction with supercomputing customers. Initial shipments to Los Alamos National Lab and the Swiss National Supercomputing Center took place in the third quarter. The UK government announced it will build one of the world's fastest AI supercomputers called Isambard-AI with almost 5,500 Grace Hopper Superchips. German supercomputing center, Julich, also announced that it will build its next-generation AI supercomputer with close to 24,000 Grace Hopper Superchips and Quantum-2 InfiniBand, making it the world's most powerful AI supercomputer with over 90 exaflops of AI performance. All-in, we estimate that the combined AI compute capacity of all the supercomputers built on Grace Hopper across the U.S., Europe and Japan next year will exceed 200 exaflops with more wins to come. Inference is contributing significantly to our data center demand, as AI is now in full production for deep learning, recommenders, chatbots, copilots and text to image generation and this is just the beginning. NVIDIA AI offers the best inference performance and versatility, and thus the lower power and cost of ownership. We are also driving a fast cost reduction curve. With the release of TensorRT-LLM, we now achieved more than 2x the inference performance for half the cost of inferencing LLMs on NVIDIA GPUs. We also announced the latest member of the Hopper family, the H200, which will be the first GPU to offer HBM3e, faster, larger memory to further accelerate generative AI and LLMs. It moves inference speed up to another 2x compared to H100 GPUs for running LLMs like Norma2 (ph). Combined, TensorRT-LLM and H200, increased performance or reduced cost by 4x in just one year. With our customers changing their stack, this is a benefit of CUDA and our architecture compatibility. Compared to the A100, H200 delivers an 18x performance increase for inferencing models like GPT-3, allowing customers to move to larger models and with no increase in latency. Amazon Web Services, Google Cloud, Microsoft Azure and Oracle Cloud will be among the first CSPs to offer H200-based instances starting next year. At last week's Microsoft Ignite, we deepened and expanded our collaboration with Microsoft across the entire stock. We introduced an AI foundry service for the development and tuning of custom generative AI enterprise applications running on Azure. Customers can bring their domain knowledge and proprietary data and we help them build their AI models using our AI expertise and software stock in our DGX cloud, all with enterprise grade security and support. SAP and Amdocs are the first customers of the NVIDIA AI foundry service on Microsoft Azure. In addition, Microsoft will launch new confidential computing instances based on the H100. The H100 remains the top performing and most versatile platform for AI training and by a wide margin, as shown in the latest MLPerf industry benchmark results. Our training cluster included more than 10,000 H100 GPUs or 3x more than in June, reflecting very efficient scaling. Efficient scaling is a key requirement in generative AI, because LLMs are growing by an order of magnitude every year. Microsoft Azure achieved similar results on a nearly identical cluster, demonstrating the efficiency of NVIDIA AI in public cloud deployments. Networking now exceeds a $10 billion annualized revenue run rate. Strong growth was driven by exceptional demand for InfiniBand, which grew fivefold year-on-year. InfiniBand is critical to gaining the scale and performance needed for training LLMs. Microsoft made this very point last week, highlighting that Azure uses over 29,000 miles of InfiniBand cabling, enough to circle the globe. We are expanding NVIDIA networking into the Ethernet space. Our new Spectrum-X end-to-end Ethernet offering with technologies, purpose built for AI, will be available in Q1 next year. With support from leading OEMs, including Dell, HPE and Lenovo. Spectrum-X can achieve 1.6x higher networking performance for AI communication compared to traditional Ethernet offerings. Let me also provide an update on our software and services offerings, where we are starting to see excellent adoption. We are on track to exit the year at an annualized revenue run rate of $1 billion for our recurring software, support and services offerings. We see two primary opportunities for growth over the intermediate term with our DGX cloud service and with our NVIDIA AI Enterprise software, each reflects the growth of enterprise AI training and enterprise AI inference, respectively. Our latest DGX cloud customer announcement was this morning as part of an AI research collaboration with Gentech, the biotechnology pioneer also plans to use our BioNeMo LLM framework to help accelerate and optimize their AI drug discovery platform. We now have enterprise AI partnership with Adobe, Dropbox, Getty, SAP, ServiceNow, Snowflake and others to come. Okay. Moving to Gaming. Gaming revenue of $2.86 billion was up 15% sequentially and up more than 80% year-on-year with strong demand in the important back-to-school shopping season with NVIDIA RTX ray tracing and AI technology now available at price points as low as $299. We entered the holidays with the best-ever line-up for gamers and creators. Gaming has doubled relative to pre-COVID levels even against the backdrop of lackluster PC market performance. This reflects the significant value we've brought to the gaming ecosystem with innovations like RTX and DLSS. The number of games and applications supporting these technologies has exploded in that period, driving upgrades and attracting new buyers. The RTX ecosystem continues to grow. There are now over 475 RTX-enabled games and applications. Generative AI is quickly emerging as the new pillar app for high performance PCs. NVIDIA RTX GPUs to find the most performance AI PCs and workstations. We just released TensorRT-LLM for Windows, which speeds on-device LLM inference up by 4x. With an installed base of over 100 million, NVIDIA RTX is the natural platform for AI application developers. Finally, our GeForce NOW cloud gaming service continues to build momentum. Its library of PC games surpassed 1,700 titles, including the launches of Alan Wake 2, Baldur's Gate 3, Cyberpunk 2077

Vivek Arya:

Thanks for taking my question. Just, Colette, wanted to clarify what China contributions are you expecting in Q4. And then, Jensen, the main question is for you, where do you think we are in the adoption curve in terms of your shipments into the generative AI market? Because when I just look at the trajectory of your data center, is growth -- it will be close to nearly 30% of all the spending in data center next year. So what metrics are you keeping an eye on to inform you that you can continue to grow? Just where are we in the adoption curve of your products into the generative AI market? Thank you.

Jensen Huang:

Generative AI is the largest TAM expansion of software and hardware that we've seen in several decades. At the core of it, what's really exciting is that, what was largely a retrieval based computing approach, almost everything that you do is retrieved off of storage somewhere, has been augmented now, added with a generative method. And it's changed almost everything. You could see that text-to-text, text-to-image, text-to-video, text-to-3D, text-to-protein, text-to-chemicals, these were things that were processed and typed in by humans in the past. And these are now generative approaches. The way that we access data is changed. It used to be based on explicit queries. It is now based on natural language queries, intention queries, semantic queries. And so, we're excited about the work that we're doing with SAP and Dropbox and many others that you're going to hear about. And one of the areas that is really impactful is the software industry, which is about $1 trillion or so, has been building tools that are manually used over the last couple of decades. And now there's a whole new segment of software called copilots and assistants. Instead of manually used, these tools will have copilots to help you use it. And so, instead of licensing software, we will continue to do that, of course, but we will also hire copilots and assistants to help us use these -- use the software. We'll connect all of these copilots and assistants into teams of AIs, which is going to be the modern version of software, modern version of enterprise business software. And so the transformation of software and the way that software has done is driving the hardware underneath. And you can see that it's transforming hardware in two ways. One is something that's largely independent of generative AI. There's two trends

Aaron Rakers:

Yeah. Thanks for taking the question. I wanted to ask about kind of the networking side of the business. Given the growth rates that you've now cited, I think, it's 155% year-over-year and strong growth sequentially, it looks like that business is like almost approaching $2.5 billion to $3 billion quarterly level. I'm curious of how you see Ethernet involved evolving and maybe how you would characterize your differentiation of Spectrum-X relative to the traditional Ethernet stack as we start to think about that becoming part of the networking narrative above and maybe beyond just InfiniBand as we look into next year? Thank you.

Operator:

Thank you. Your next question comes from the line of Joe Moore of Morgan Stanley. Your line is open.

Jensen Huang:

Yeah. Thanks for the question. Grace Hopper is in production -- in high volume production now. We're expecting next year just with all of the design wins that we have in high performance computing and AI infrastructures, we are on a very, very fast ramp with our first data center CPU to a multi-billion dollar product line. This is going to be a very large product line for us. The capability of Grace Hopper is really quite spectacular. It has the ability to create computing nodes that simultaneously has very fast memory, as well as very large memory. In the areas of vector databases or semantic surge, what is called RAG, retrieval augmented generation. So that you could have a generative AI model be able to refer to proprietary data or a factual data before it generates a response, that data is quite large. And you can also have applications or generative models where the context length is very high. You basically store it in entire book into end-to-end system memory before you ask your questions. And so the context length can be quite large this way. The generative models has the ability to still be able to naturally interact with you on one hand. On the other hand, be able to refer to factual data, proprietary data or domain-specific data, you data and be contextually relevant and reduce hallucination. And so that particular use case for example is really quite fantastic for Grace Hopper. It also serves the customers that really care to have a different CPU than x86. Maybe it's a European supercomputing centers or European companies who would like to build up their own ARM ecosystem and like to build up a full stack or CSPs that have decided that they would like to pivot to ARM, because their own custom CPUs are based on ARM. There are variety of different reasons that drives the success of Grace Hopper, but we're off to a just an extraordinary start. This is a home run product.

Tim Arcuri:

Hi. Thanks. I wanted to ask a little bit about the visibility that you have on revenue. I know there's a few moving parts. I guess, on one hand, the purchase commitments went up a lot again. But on the other hand, China bans would arguably pull in when you can fill the demand beyond China. So I know we're not even into 2024 yet and it doesn't sound like, Jensen, you think that next year would be a peak in your Data Center revenue, but I just wanted to sort of explicitly ask you that. Do you think that Data Center can grow even in 2025? Thanks.

Operator:

Your next question comes from the line of Toshiya Hari of Goldman Sachs. Your line is open.

Colette Kress:

Thanks, Toshiya. On the question regarding potentially new products that we could provide to our China customers. It's a significant process to both design and develop these new products. As we discussed, we're going to make sure that we are in full discussions with the U.S. government of our intent to move products as well. Given our state about where we are in the quarter, we're already several weeks into the quarter. So it's just going to take some time for us to go through and discussing with our customers the needs and desires of these new products that we have. And moving forward, whether that's medium-term or long-term, it's just hard to say both the [Technical Difficulty] of what we can produce with the U.S. government and what the interest of our China customers in this. So we stay still focused on finding that right balance for our China customers, but it's hard to say at this time.

Operator:

Your next question comes from the line of Stacy Rasgon of Bernstein Research. Your line is open.

Colette Kress:

Yeah. Stacy, let me see if I can help you understand. Yes, there are still situations where we are working on both improving our supply each and every quarter. We've done a really solid job of ramping every quarter, which has defined our revenue. But with the absence of China for our outlook for Q4, sure, there could have been some things that we are not supply-constrained that we could have sold, but kind of we would no longer can. So could our guidance had been a little higher in our Q4? Yes. We are still working on improving our supply on plan, on continuing growing all throughout next year as well towards that.

Matt Ramsay:

Thank you very much. Congrats, everybody, on the results. Jensen, I had a two-part question for you, and it comes off of sort of one premise. And the premise is, I still get a lot of questions from investors thinking about AI training as being NVIDIA's dominant domain and somehow inference, even large model inference takes more and more of the TAM that the market will become more competitive. You'll be less differentiated et cetera., et cetera. So I guess the two parts of the question are

Operator:

Your final question comes from the line of Harlan Sur of J.P. Morgan. Your line is open.

Jensen Huang:

Gosh. Boy, that's just really excellent. You just wrote NVIDIA's business plan, and so you described our strategy. First of all, there is a fundamental reason why we accelerate our execution. And the reason for that is because it fundamentally drives down cost. When the combination of TensorRT-LLM and H200 reduce the cost for our customers for large model inference by a factor of four, and so that includes, of course, our speeds and feeds, but mostly it's because of our software, mostly the software benefits because of the architecture. And so we want to accelerate our roadmap for that reason. The second reason is to expand the reach of generative AI, the world's number of data center configurations -- this is kind of the amazing thing. NVIDIA is in every cloud, but not one cloud is the same. NVIDIA is working with every single cloud service provider and not one of the networking control plane, security posture is the same. Everybody's platform is different and yet we're integrated into all of their stacks, all of their data centers and we work incredibly well with all of them. And not to mention, we then take the whole thing and we create AI factories that are standalone. We take our platform, we can put them into supercomputers, we can put them into enterprise. Bringing AI to enterprise is something generative AI Enterprise something nobody's ever done before. And we're right now in the process of going to market with all of that. And so the complexity includes, of course, all the technologies and segments and the pace. It includes the fact that we are architecturally compatible across every single one of those. It includes all of the domain specific libraries that we create. The reason why every computer company, without thinking, can integrate NVIDIA into their roadmap and take it to market. And the reason for that is, because there is market demand for it. There is market demand in healthcare, there is market demand in manufacturing, there is market demand, of course, in AI, including financial services, in supercomputing and quantum computing. The list of markets and segments that we have domain specific libraries is incredibly broaden. And then finally, we have an end-to-end solution for data centers; InfiniBand networking, Ethernet networking, x86, ARM, just about every permutation combination of solutions -- technology solutions and software stacks provided. And that translates to having the largest number of ecosystem software developers; the largest ecosystem of system makers; the largest and broadest distribution partnership network; and ultimately, the greatest reach. And that takes -- surely that takes a lot of energy. But the thing that really holds it together, and this is a great decision that we made decades ago, which is everything is architecturally compatible. When we develop a domain specific language that runs on one GPU, it runs on every GPU. When we optimize TensorRT for the cloud, we optimized it for enterprise. When we do something that brings in a new feature, a new library, a new feature or a new developer, they instantly get the benefit of all of our reach. And so that discipline, that architecture compatible discipline that has lasted more than a couple of decades now, is one of the reasons why NVIDIA is still really, really efficient. I mean, we're 28,000 people large and serving just about every single company, every single industry, every single market around the world.

Jensen Huang:

Our strong growth reflects the broad industry platform transition from general purpose to accelerated computing and generative AI. Large language models start-ups consumer Internet companies and global cloud service providers are the first movers. The next waves are starting to build. Nations and regional CSPs are building AI clouds to serve local demand. Enterprise software companies like Adobe and Dropbox, SAP and ServiceNow are adding AI copilots and assistants to their platforms. Enterprises in the world's largest industries are creating custom AIs to automate and boost productivity. The generative AI era is in full steam and has created the need for a new type of data center, an AI factory; optimized for refining data and training, and inference, and generating AI. AI factory workloads are different and incremental to legacy data center workloads supporting IT tasks. AI factories run copilots and AI assistants, which are significant software TAM expansion and are driving significant new investment. Expanding the $1 trillion traditional data center infrastructure installed base, empowering the AI Industrial Revolution. NVIDIA H100 HGX with InfiniBand and the NVIDIA AI software stack define an AI factory today. As we expand our supply chain to meet the world's demand, we are also building new growth drivers for the next wave of AI. We highlighted three elements to our new growth strategy that are hitting their stride

Operator:

Good afternoon. My name is David, and I'll be your conference operator today. At this time, I'd like to welcome everyone to NVIDIA's Second Quarter Earnings Call. Today's conference is being recorded. All lines have been placed on mute to prevent any background noise. After the speakers’ remarks, there will be a question-and-answer session. [Operator Instructions] Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. We had an exceptional quarter. Record Q2 revenue of $13.51 billion was up 88% sequentially and up 101% year-on-year, and above our outlook of $11 billion. Let me first start with Data Center. Record revenue of $10.32 billion was up 141% sequentially and up 171% year-on-year. Data Center compute revenue nearly tripled year-on-year, driven primarily by accelerating demand from cloud service providers and large consumer Internet companies for HGX platform, the engine of generative AI and large language models. Major companies, including AWS, Google Cloud, Meta, Microsoft Azure and Oracle Cloud as well as a growing number of GPU cloud providers are deploying, in volume, HGX systems based on our Hopper and Ampere architecture Tensor Core GPUs. Networking revenue almost doubled year-on-year, driven by our end-to-end InfiniBand networking platform, the gold standard for AI. There is tremendous demand for NVIDIA accelerated computing and AI platforms. Our supply partners have been exceptional in ramping capacity to support our needs. Our data center supply chain, including HGX with 35,000 parts and highly complex networking has been built up over the past decade. We have also developed and qualified additional capacity and suppliers for key steps in the manufacturing process such as [indiscernible] packaging. We expect supply to increase each quarter through next year. By geography, data center growth was strongest in the U.S. as customers direct their capital investments to AI and accelerated computing. China demand was within the historical range of 20% to 25% of our Data Center revenue, including compute and networking solutions. At this time, let me take a moment to address recent reports on the potential for increased regulations on our exports to China. We believe the current regulation is achieving the intended results. Given the strength of demand for our products worldwide, we do not anticipate that additional export restrictions on our Data Center GPUs, if adopted, would have an immediate material impact to our financial results. However, over the long term, restrictions prohibiting the sale of our Data Center GPUs to China, if implemented, will result in a permanent loss and opportunity for the U.S. industry to compete and lead in one of the world's largest markets. Our cloud service providers drove exceptional strong demand for HGX systems in the quarter, as they undertake a generational transition to upgrade their data center infrastructure for the new era of accelerated computing and AI. The NVIDIA HGX platform is culminating of nearly two decades of full stack innovation across silicon, systems, interconnects, networking, software and algorithms. Instances powered by the NVIDIA H100 Tensor Core GPUs are now generally available at AWS, Microsoft Azure and several GPU cloud providers, with others on the way shortly. Consumer Internet companies also drove the very strong demand. Their investments in data center infrastructure purpose-built for AI are already generating significant returns. For example, Meta, recently highlighted that since launching Reels, AI recommendations have driven a more than 24% increase in time spent on Instagram. Enterprises are also racing to deploy generative AI, driving strong consumption of NVIDIA powered instances in the cloud as well as demand for on-premise infrastructure. Whether we serve customers in the cloud or on-prem through partners or direct, their applications can run seamlessly on NVIDIA AI enterprise software with access to our acceleration libraries, pre-trained models and APIs. We announced a partnership with Snowflake to provide enterprises with accelerated path to create customized generative AI applications using their own proprietary data, all securely within the Snowflake Data Cloud. With the NVIDIA NeMo platform for developing large language models, enterprises will be able to make custom LLMs for advanced AI services, including chatbot, search and summarization, right from the Snowflake Data Cloud. Virtually, every industry can benefit from generative AI. For example, AI Copilot such as those just announced by Microsoft can boost the productivity of over 1 billion office workers and tens of millions of software engineers. Billions of professionals in legal services, sales, customer support and education will be available to leverage AI systems trained in their field. AI Copilot and assistants are set to create new multi-hundred billion dollar market opportunities for our customers. We are seeing some of the earliest applications of generative AI in marketing, media and entertainment. WPP, the world's largest marketing and communication services organization, is developing a content engine using NVIDIA Omniverse to enable artists and designers to integrate generative AI into 3D content creation. WPP designers can create images from text prompts while responsibly trained generative AI tools and content from NVIDIA partners such as Adobe and Getty Images using NVIDIA Picasso, a foundry for custom generative AI models for visual design. Visual content provider Shutterstock is also using NVIDIA Picasso to build tools and services that enables users to create 3D scene background with the help of generative AI. We've partnered with ServiceNow and Accenture to launch the AI Lighthouse program, fast tracking the development of enterprise AI capabilities. AI Lighthouse unites the ServiceNow enterprise automation platform and engine with NVIDIA accelerated computing and with Accenture consulting and deployment services. We are collaborating also with Hugging Face to simplify the creation of new and custom AI models for enterprises. Hugging Face will offer a new service for enterprises to train and tune advanced AI models powered by NVIDIA HGX cloud. And just yesterday, VMware and NVIDIA announced a major new enterprise offering called VMware Private AI Foundation with NVIDIA, a fully integrated platform featuring AI software and accelerated computing from NVIDIA with multi-cloud software for enterprises running VMware. VMware's hundreds of thousands of enterprise customers will have access to the infrastructure, AI and cloud management software needed to customize models and run generative AI applications such as intelligent chatbot, assistants, search and summarization. We also announced new NVIDIA AI enterprise-ready servers featuring the new NVIDIA L40S GPU built for the industry standard data center server ecosystem and BlueField-3 DPU data center infrastructure processor. L40S is not limited by [indiscernible] supply and is shipping to the world's leading server system makers (ph). L40S is a universal data center processor designed for high volume data center standing out to accelerate the most compute-intensive applications, including AI training and inventing through the designing, visualization, video processing and NVIDIA Omniverse industrial digitalization. NVIDIA AI enterprise ready servers are fully optimized for VMware, Cloud Foundation and Private AI Foundation. Nearly 100 configurations of NVIDIA AI enterprise ready servers will soon be available from the world's leading enterprise IT computing companies, including Dell, HP and Lenovo. The GH200 Grace Hopper Superchip which combines our ARM-based Grace CPU with Hopper GPU entered full production and will be available this quarter in OEM servers. It is also shipping to multiple supercomputing customers, including Atmos (ph), National Labs and the Swiss National Computing Center. And NVIDIA and SoftBank are collaborating on a platform based on GH200 for generative AI and 5G/6G applications. The second generation version of our Grace Hopper Superchip with the latest HBM3e memory will be available in Q2 of calendar 2024. We announced the DGX GH200, a new class of large memory AI supercomputer for giant AI language model, recommendator systems and data analytics. This is the first use of the new NVIDIA [indiscernible] switch system, enabling all of its 256 Grace Hopper Superchips to work together as one, a huge jump compared to our prior generation connecting just eight GPUs over [indiscernible]. DGX GH200 systems are expected to be available by the end of the year, Google Cloud, Meta and Microsoft among the first to gain access. Strong networking growth was driven primarily by InfiniBand infrastructure to connect HGX GPU systems. Thanks to its end-to-end optimization and in-network computing capabilities, InfiniBand delivers more than double the performance of traditional Ethernet for AI. For billions of dollar AI infrastructures, the value from the increased throughput of InfiniBand is worth hundreds of [indiscernible] and pays for the network. In addition, only InfiniBand can scale to hundreds of thousands of GPUs. It is the network of choice for leading AI practitioners. For Ethernet-based cloud data centers that seek to optimize their AI performance, we announced NVIDIA Spectrum-X, an accelerated networking platform designed to optimize Ethernet for AI workloads. Spectrum-X couples the Spectrum or Ethernet switch with the BlueField-3 DPU, achieving 1.5x better overall AI performance and power efficiency versus traditional Ethernet. BlueField-3 DPU is a major success. It is in qualification with major OEMs and ramping across multiple CSPs and consumer Internet companies. Now moving to gaming. Gaming revenue of $2.49 billion was up 11% sequentially and 22% year-on-year. Growth was fueled by GeForce RTX 40 Series GPUs for laptops and desktop. End customer demand was solid and consistent with seasonality. We believe global end demand has returned to growth after last year's slowdown. We have a large upgrade opportunity ahead of us. Just 47% of our installed base have upgraded to RTX and about 20% of the GPU with an RTX 3060 or higher performance. Laptop GPUs posted strong growth in the key back-to-school season, led by RTX 4060 GPUs. NVIDIA's GPU-powered laptops have gained in popularity, and their shipments are now outpacing desktop GPUs from several regions around the world. This is likely to shift the reality of our overall gaming revenue a bit, with Q2 and Q3 as the stronger quarters of the year, reflecting the back-to-school and holiday build schedules for laptops. In desktop, we launched the GeForce RTX 4060 and the GeForce RTX 4060 TI GPUs, bringing the Ada Lovelace architecture down to price points as low as $299. The ecosystem of RTX and DLSS games continue to expand. 35 new games added to DLSS support, including blockbusters such as Diablo IV and Baldur’s Gate 3. There's now over 330 RTX accelerated games and apps. We are bringing generative AI to gaming. At COMPUTEX, we announced NVIDIA Avatar Cloud Engine or ACE for games, a custom AI model foundry service. Developers can use this service to bring intelligence to non-player characters. And it harnesses a number of NVIDIA Omniverse and AI technologies, including NeMo, Riva and Audio2Face. Now moving to Professional Visualization. Revenue of $375 million was up 28% sequentially and down 24% year-on-year. The Ada architecture ramp drove strong growth in Q2, rolling out initially in laptop workstations with a refresh of desktop workstations coming in Q3. These will include powerful new RTX systems with up to 4 NVIDIA RTX 6000 GPUs, providing more than 5,800 teraflops of AI performance and 192 gigabytes of GPU memory. They can be configured with NVIDIA AI enterprise or NVIDIA Omniverse inside. We also announced three new desktop workstation GPUs based on the Ada generation. The NVIDIA RTX 5000, 4500 and 4000, offering up to 2x the RT core throughput and up to 2x faster AI training performance compared to the previous generation. In addition to traditional workloads such as 3D design and content creation, new workloads in generative AI, large language model development and data science are expanding the opportunity in pro visualization for our RTX technology. One of the key themes in Jensen's keynote [indiscernible] earlier this month was the conversion of graphics and AI. This is where NVIDIA Omniverse is positioned. Omniverse is OpenUSD's native platform. OpenUSD is a universal interchange that is quickly becoming the standard for the 3D world, much like HTML is the universal language for the 2D [indiscernible]. Together, Adobe, Apple, Autodesk, Pixar and NVIDIA form the Alliance for OpenUSD. Our mission is to accelerate OpenUSD's development and adoption. We announced new and upcoming Omniverse cloud APIs, including RunUSD and ChatUSD to bring generative AI to OpenUSD workload. Moving to automotive. Revenue was $253 million, down 15% sequentially and up 15% year-on-year. Solid year-on-year growth was driven by the ramp of self-driving platforms based on [indiscernible] or associated with a number of new energy vehicle makers. The sequential decline reflects lower overall automotive demand, particularly in China. We announced a partnership with MediaTek to bring drivers and passengers new experiences inside the car. MediaTek will develop automotive SoCs and integrate a new product line of NVIDIA's GPU chiplet. The partnership covers a wide range of vehicle segments from luxury to entry level. Moving to the rest of the P&L. GAAP gross margins expanded to 70.1% and non-GAAP gross margin to 71.2%, driven by higher data center sales. Our Data Center products include a significant amount of software and complexity, which is also helping drive our gross margin. Sequential GAAP operating expenses were up 6% and non-GAAP operating expenses were up 5%, primarily reflecting increased compensation and benefits. We returned approximately $3.4 billion to shareholders in the form of share repurchases and cash dividends. Our Board of Directors has just approved an additional $25 billion in stock repurchases to add to our remaining $4 billion of authorization as of the end of Q2. Let me turn to the outlook for the third quarter of fiscal 2024. Demand for our Data Center platform where AI is tremendous and broad-based across industries on customers. Our demand visibility extends into next year. Our supply over the next several quarters will continue to ramp as we lower cycle times and work with our supply partners to add capacity. Additionally, the new L40S GPU will help address the growing demand for many types of workloads from cloud to enterprise. For Q3, total revenue is expected to be $16 billion, plus or minus 2%. We expect sequential growth to be driven largely by Data Center with gaming and ProViz also contributing. GAAP and non-GAAP gross margins are expected to be 71.5% and 72.5%, respectively, plus or minus 50 basis points. GAAP and non-GAAP operating expenses are expected to be approximately $2.95 billion and $2 billion, respectively. GAAP and non-GAAP other income and expenses are expected to be an income of approximately $100 million, excluding gains and losses from non-affiliated investments. GAAP and non-GAAP tax rates are expected to be 14.5%, plus or minus 1%, excluding any discrete items. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight some upcoming events for the financial community. We will attend the Jefferies Tech Summit on August 30 in Chicago, the Goldman Sachs Conference on September 5 in San Francisco, the Evercore Semiconductor Conference on September 6 as well as the Citi Tech Conference on September 7, both in New York. And the BofA Virtual AI conference on September 11. Our earnings call to discuss the results of our third quarter of fiscal 2024 is scheduled for Tuesday, November 21. Operator, we will now open the call for questions. Could you please poll for questions for us? Thank you.

Matt Ramsay:

Yes. Thank you very much. Good afternoon. Obviously, remarkable results. Jensen, I wanted to ask a question of you regarding the really quickly emerging application of large model inference. So I think it's pretty well understood by the majority of investors that you guys have very much a lockdown share of the training market. A lot of the smaller market -- smaller model inference workloads have been done on ASICs or CPUs in the past. And with many of these GPT and other really large models, there's this new workload that's accelerating super-duper quickly on large model inference. And I think your Grace Hopper Superchip products and others are pretty well aligned for that. But could you maybe talk to us about how you're seeing the inference market segment between small model inference and large model inference and how your product portfolio is positioned for that? Thanks.

Operator:

Okay. Next, we'll go to Vivek Arya with BofA Securities. Your line is now open.

Colette Kress:

So thanks for that question regarding our supply. Yes, we do expect to continue increasing ramping our supply over the next quarters as well as into next fiscal year. In terms of percent, it's not something that we have here. It is a work across so many different suppliers, so many different parts of building an HGX and many of our other new products that are coming to market. But we are very pleased with both the support that we have with our suppliers and the long time that we have spent with them improving their supply.

Operator:

Next, we go to Stacy Rasgon with Bernstein Research. Your line is open.

Colette Kress:

Sure, Stacy. Let me help. Within the quarter, our HGX systems were a very significant part of our Data Center as well as our Data Center growth that we had seen. Those systems include our HGX of our Hopper architecture, but also our Ampere architecture. Yes, we are still selling both of these architectures in the market. Now when you think about that, what does that mean from both the systems as a unit, of course, is growing quite substantially, and that is driving in terms of the revenue increases. So both of these things are the drivers of the revenue inside Data Center. Our DGXs are always a portion of additional systems that we will sell. Those are great opportunities for enterprise customers and many other different types of customers that we're seeing even in our consumer Internet companies. The importance there is also coming together with software that we sell with our DGXs, but that's a portion of our sales that we're doing. The rest of the GPUs, we have new GPUs coming to market that we talk about the L40S, and they will add continued growth going forward. But again, the largest driver of our revenue within this last quarter was definitely the HGX system.

Operator:

Next, we go to Mark Lipacis with Jefferies. Your line is now open.

A – Jensen Huang:

Yeah, Mark, I really appreciate the question. Let me see if I could use some metrics, so we have a run time called AI Enterprise. This is one part of our software stack. And this is, if you will, the run time that just about every company uses for the end-to-end of machine learning from data processing, the training of any model that you like to do on any framework you'd like to do, the inference and the deployment, the scaling it out into a data center. It could be a scale-out for a hyperscale data center. It could be a scale-out for enterprise data center, for example, on VMware. You can do this on any of our GPUs. We have hundreds of millions of GPUs in the field and millions of GPUs in the cloud and just about every single cloud. And it runs in a single GPU configuration as well as multi-GPU per compute or multi-node. It also has multiple sessions or multiple computing instances per GPU. So from multiple instances per GPU to multiple GPUs, multiple nodes to entire data center scale. So this run time called NVIDIA AI enterprise has something like 4,500 software packages, software libraries and has something like 10,000 dependencies among each other. And that run time is, as I mentioned, continuously updated and optimized for our installed base for our stack. And that's just one example of what it would take to get accelerated computing to work. The number of code combinations and type of application combinations is really quite insane. And it's taken us two decades to get here. But what I would characterize as probably our -- the elements of our company, if you will, are several. I would say number 1 is architecture. The flexibility, the versatility and the performance of our architecture makes it possible for us to do all the things that I just said, from data processing to training to inference, for preprocessing of the data before you do the inference to the post processing of the data, tokenizing of languages so that you could then train with it. The amount of -- the workflow is much more intense than just training or inference. But anyways, that's where we'll focus and it's fine. But when people actually use these computing systems, it's quite -- requires a lot of applications. And so the combination of our architecture makes it possible for us to deliver the lowest cost ownership. And the reason for that is because we accelerate so many different things. The second characteristic of our company is the installed base. You have to ask yourself, why is it that all the software developers come to our platform? And the reason for that is because software developers seek a large installed base so that they can reach the largest number of end users, so that they could build a business or get a return on the investments that they make. And then the third characteristic is reach. We're in the cloud today, both for public cloud, public-facing cloud because we have so many customers that use -- so many developers and customers that use our platform. CSPs are delighted to put it up in the cloud. They use it for internal consumption to develop and train and to operate recommender systems or search or data processing engines and whatnot all the way to training and inference. And so we're in the cloud, we're in enterprise. Yesterday, we had a very big announcement. It's really worthwhile to take a look at that. VMware is the operating system of the world's enterprise. And we've been working together for several years now, and we're going to bring together -- together, we're going to bring generative AI to the world's enterprises all the way out to the edge. And so reach is another reason. And because of reach, all of the world's system makers are anxious to put NVIDIA's platform in their systems. And so we have a very broad distribution from all of the world's OEMs and ODMs and so on and so forth because of our reach. And then lastly, because of our scale and velocity, we were able to sustain this really complex stack of software and hardware, networking and compute and across all of these different usage models and different computing environments. And we're able to do all this while accelerating the velocity of our engineering. It seems like we're introducing a new architecture every two years. Now we're introducing a new architecture, a new product just about every six months. And so these properties make it possible for the ecosystem to build their company and their business on top of us. And so those in combination makes us special.

Atif Malik:

Hi. Thank you for taking my question. Great job on results and outlook. Colette, I have a question on the core L40S that you guys talked about. Any idea how much of the supply tightness can L40S help with? And if you can talk about the incremental profitability or gross margin contribution from this product? Thank you.

Operator:

Next, we'll go to Joe Moore with Morgan Stanley. Your line is open.

Jensen Huang:

Yeah. We have excellent visibility through the year and into next year. And we're already planning the next-generation infrastructure with the leading CSPs and data center builders. The demand – easiest way to think about the demand, the world is transitioning from general-purpose computing to accelerated computing. That's the easiest way to think about the demand. The best way for companies to increase their throughput, improve their energy efficiency, improve their cost efficiency is to divert their capital budget to accelerated computing and generative AI. Because by doing that, you're going to offload so much workload off of the CPUs, but the available CPUs is -- in your data center will get boosted. And so what you're seeing companies do now is recognizing this -- the tipping point here, recognizing the beginning of this transition and diverting their capital investment to accelerated computing and generative AI. And so that's probably the easiest way to think about the opportunity ahead of us. This isn't a singular application that is driving the demand, but this is a new computing platform, if you will, a new computing transition that's happening. And data centers all over the world are responding to this and shifting in a broad-based way.

Toshiya Hari:

Hi. Thank you for taking the question. I had one quick clarification question for Colette and then another one for Jensen. Colette, I think last quarter, you had said CSPs were about 40% of your Data Center revenue, consumer Internet at 30%, enterprise 30%. Based on your remarks, it sounded like CSPs and consumer Internet may have been a larger percentage of your business. If you can kind of clarify that or confirm that, that would be super helpful. And then Jensen, a question for you. Given your position as the key enabler of AI, the breadth of engagements and the visibility you have into customer projects, I'm curious how confident you are that there will be enough applications or use cases for your customers to generate a reasonable return on their investments. I guess I ask the question because there is a concern out there that there could be a bit of a pause in your demand profile in the out years. Curious if there's enough breadth and depth there to support a sustained increase in your Data Center business going forward. Thank you.

Jensen Huang:

Toshi, I'm reluctant to guess about the future and so I'll answer the question from the first principle of computer science perspective. It is recognized for some time now that general purpose computing is just not and brute forcing general purpose computing. Using general purpose computing at scale is no longer the best way to go forward. It's too energy costly, it's too expensive, and the performance of the applications are too slow. And finally, the world has a new way of doing it. It's called accelerated computing and what kicked it into turbocharge is generative AI. But accelerated computing could be used for all kinds of different applications that's already in the data center. And by using it, you offload the CPUs. You save a ton of money in order of magnitude, in cost and order of magnitude and energy and the throughput is higher and that's what the industry is really responding to. Going forward, the best way to invest in the data center is to divert the capital investment from general purpose computing and focus it on generative AI and accelerated computing. Generative AI provides a new way of generating productivity, a new way of generating new services to offer to your customers, and accelerated computing helps you save money and save power. And the number of applications is, well, tons. Lots of developers, lots of applications, lots of libraries. It's ready to be deployed. And so I think the data centers around the world recognize this, that this is the best way to deploy resources, deploy capital going forward for data centers. This is true for the world's clouds and you're seeing a whole crop of new GPU specialty -- GPU specialized cloud service providers. One of the famous ones is CoreWeave and they're doing incredibly well. But you're seeing the regional GPU specialist service providers all over the world now. And it's because they all recognize the same thing, that the best way to invest their capital going forward is to put it into accelerated computing and generative AI. We're also seeing that enterprises want to do that. But in order for enterprises to do it, you have to support the management system, the operating system, the security and software-defined data center approach of enterprises, and that's all VMware. And we've been working several years with VMware to make it possible for VMware to support not just the virtualization of CPUs but a virtualization of GPUs as well as the distributed computing capabilities of GPUs, supporting NVIDIA's BlueField for high-performance networking. And all of the generative AI libraries that we've been working on is now going to be offered as a special SKU by VMware's sales force, which is, as we all know, quite large because they reach some several hundred thousand VMware customers around the world. And this new SKU is going to be called VMware Private AI Foundation. And this will be a new SKU that makes it possible for enterprises. And in combination with HP, Dell, and Lenovo's new server offerings based on L40S, any enterprise could have a state-of-the-art AI data center and be able to engage generative AI. And so I think the answer to that question is hard to predict exactly what's going to happen quarter-to-quarter. But I think the trend is very, very clear now that we're seeing a platform shift.

Timothy Arcuri:

Thanks a lot. Can you talk about the attach rate of your networking solutions to your -- to the compute that you're shipping? In other words, is like half of your compute shipping with your networking solutions more than half, less than half? And is this something that maybe you can use to prioritize allocation of the GPUs? Thank you.

Operator:

Our final question comes from the line of Ben Reitzes with Melius. Your line is now open.

Jensen Huang:

DGX Cloud's strategy, let me start there. DGX Cloud's strategy is to achieve several things

Operator:

And that does conclude today's question-and-answer session. I'll turn the call back over to Jensen Huang for any additional or closing remarks.

Operator:

This concludes today's conference call. You may now disconnect.

Operator:

Good afternoon. My name is David, and I'll be your conference operator today. At this time, I'd like to welcome everyone to NVIDIA's First Quarter Earnings Call. Today's conference is being recorded. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there'll be a question-and-answer session. [Operator Instructions] Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. Q1 revenue was $7.19 billion, up 19% sequentially and down 13% year-on-year. Strong sequential growth was driven by record data center revenue, with our gaming and professional visualization platforms emerging from channel inventory corrections. Starting with data center, record revenue of $4.28 billion was up 18% sequentially and up 14% year-on-year, on strong growth by accelerated computing platform worldwide. Generative AI is driving exponential growth in compute requirements and a fast transition to NVIDIA accelerated computing, which is the most versatile, most energy-efficient, and the lowest TCO approach to train and deploy AI. Generative AI drove significant upside in demand for our products, creating opportunities and broad-based global growth across our markets. Let me give you some color across our three major customer categories, cloud service providers or CSPs, consumer Internet companies, and enterprises. First, CSPs around the world are racing to deploy our flagship Hopper and Ampere architecture GPUs to meet the surge in interest from both enterprise and consumer AI applications for training and inference. Multiple CSPs announced the availability of H100 on their platforms, including private previews at Microsoft Azure, Google Cloud, and Oracle Cloud Infrastructure, upcoming offerings at AWS, and general availability at emerging GPU specialized cloud providers like CoreWeave and Lambda. In addition to enterprise AI adoption, these CSPs are serving strong demand for H100 from Generative AI pioneers. Second, consumer Internet companies are also at the forefront of adopting Generative AI and deep learning-based recommendation systems, driving strong growth. For example, Meta has now deployed it's H100 powered Grand Teton AI supercomputer for its AI production and research teams. Third, enterprise demand for AI and accelerated computing is strong. We are seeing momentum in verticals such as automotive, financial services, healthcare, and telecom, where AI and accelerated computing are quickly becoming integral to customers' innovation roadmaps and competitive positioning. For example, Bloomberg announced it has a 50 billion parameter model, BloombergGPT, to help with financial natural language processing tasks such as sentiment analysis, named entity recognition, news classification, and question-answering. Auto Insurance company, CCC Intelligent Solutions is using AI for estimating repairs. And AT&T is working with us on AI to improve fleet dispatches so their field technicians can better serve customers. Among other enterprise customers using NVIDIA AI are Deloitte for logistics and customer service and Amgen for drug discovery and protein engineering. This quarter, we started shipping DGX H100, our Hopper generation AI system, which customers can deploy on-prem. And with the launch of DGX Cloud through our partnership with Microsoft Azure, Google Cloud, and Oracle Cloud Infrastructure, we deliver the promise of NVIDIA DGX to customers from the cloud. Whether the customers deploy DGX on-prem or via DGX Cloud, they get access to NVIDIA AI software, including NVIDIA Base Command, and AI frameworks, and pre-trained models. We provide them with the blueprint for building and operating AI, spanning our expertise across systems, algorithms, data processing, and training methods. We also announced NVIDIA AI Foundations, which are model foundry services available on DGX Cloud, that enable businesses to build, refine, and operate custom large language models and generative AI models, trained with our own proprietary data, created for unique domain-specific tasks. They include NVIDIA NeMo for large language models, NVIDIA Picasso for images, video, and 3D, and NVIDIA BioNeMo for life sciences. Each service has six elements, pre-trained models, frameworks for data processing and curation, proprietary knowledge-based sector databases, systems for fine-tuning, aligning, and guardrailing, optimized inference engines, and support from NVIDIA experts to help enterprises fine-tune models for their custom use cases. ServiceNow, a leading enterprise services platform is an early adopter of DGX Cloud and NeMo. They are developing custom large language models trained on data specifically for the ServiceNow platform. Our collaboration will let ServiceNow create new enterprise-grade generative AI offerings, with the 1,000s of enterprises worldwide running on the ServiceNow platform, including for IT departments, customer service teams, employees, and developers. Generative AI is also driving a step-function increase in inference workloads. Because of their size and complexities, these workloads require acceleration. The latest MLPerf industry benchmark released in April showed NVIDIA's inference platform deliver performance that is orders of magnitude ahead of the industry, with unmatched versatility across diverse workloads. To help customers deploy generative AI applications at scale, at GTC, we announced four major new inference platforms that leverage the NVIDIA AI software stack. These include L4 Tensor Core GPU for AI video, L40 for Omniverse, and graphics rendering, H100 NVL for large language models, and the Grace Hopper Superchip for LLMs and also, recommendation systems and vector databases. Google Cloud is the first CSP to adopt our L4 inference platform with the launch of its G2 virtual machines for generative AI inference and other workloads such as Google Cloud Dataproc, Google AlphaFold, and Google Cloud's Immersive Stream, which render 3D and AR experiences. In addition, Google is integrating our Triton inference server with Google Kubernetes engine and its cloud-based Vertex AI platform. In networking, we saw strong demand at both CSPs and enterprise customers for generative AI and accelerated computing, which require high-performance networking like NVIDIA's Mellanox networking platforms. Demand relating to general purpose CPU infrastructure remain soft. As generative AI applications grow in size and complexity, high performance networks become essential for delivering accelerated computing at data center scale to meet the enormous demand of all training and inferencing. Our 400 gig Quantum-2 InfiniBand platform is the gold standard for AI dedicated infrastructure, with broad adoption across major cloud and consumer Internet platforms such as Microsoft Azure. With the combination of in-network computing technology and the industry's only end-to-end data center scale, optimized software stack, customers routinely enjoy a 20% increase in throughput for their sizable infrastructure investment. For multi-tenant cloud transitioning to support generative AI our high-speed Ethernet platform with BlueField-3 DPUs and Spectrum-4 Ethernet switching, offers the highest available Ethernet network performance. BlueField-3 is in production and has been adopted by multiple hyperscale and CSP customers, including Microsoft Azure, Oracle Cloud, CoreWeave, Baidu, and others. We look forward to sharing more about our 400 gig Spectrum-4 accelerated AI networking platform next week at the COMPUTEX Conference in Taiwan. Lastly, our Grace data center CPU is sampling with customers. At this week's International Supercomputing Conference in Germany, the University of Bristol announced a new supercomputer based on the NVIDIA Grace CPU Superchip, which is 6x more energy-efficient than the previous supercomputer. This adds to the growing momentum for Grace with both CPU only and CPU/GPU opportunities across AI and cloud and supercomputing applications. The coming wave of BlueField-3, Grace and Grace Hopper Superchips will enable a new generation of super energy efficient accelerated data centers. Now, let's move to gaming. Gaming revenue of $2.24 billion was up 22% sequentially, and down 38% year-on-year. Strong sequential growth was driven by sales of the 40 Series GeForce RTX GPUs for both notebooks and desktops. Overall, end demand was solid, and consistent with seasonality, demonstrating resilience against a challenging consumer spending backdrop. The GeForce RTX 40 Series GPU laptops are off to a great start, featuring four NVIDIA inventions, RTX Path Tracing, DLSS 3 AI rendering, Reflex Ultra-Low Latency rendering, and Max-Q, energy efficient technologies. They deliver tremendous gains in industrial design, performance and battery life for gamers and creators. And like our desktop offerings, 40 Series laptops support the NVIDIA Studio platform or software technologies, including acceleration for creative data science and AI workflows, and Omniverse, giving content creators unmatched tools and capabilities. In desktop, we ramped the RTX 4070, which joined the previously launched RTX 4090, 4080, and 4070 Ti GPUs. The RTX 4070 is nearly 3x faster than the RTX 2070 and offers our large installed-base a spectacular upgrade. Last week, we launched the 60 family, RTX 4060, and 4060 Ti, bringing our newest architecture to the world's core gamers starting at just $299. These GPUs for the first time provide 2x the performance of the latest gaming console at mainstream price points. The 4060 Ti is available starting today, while the 4060 will be available in July. Generative AI will be transformative to gaming and content creation from development to run time. At the Microsoft Build Developer Conference earlier this week, we showcased how Windows PCs and workstations with NVIDIA RTX GPUs will be AI-powered at their core. NVIDIA and Microsoft have collaborated on end-to-end software engineering, spanning from the Windows operating system to the NVIDIA graphics drivers, and NeMo's LLM framework to help make Windows on NVIDIA RTX Tensor Core GPUs, a supercharged platform for generative AI. Last quarter, we announced a partnership with Microsoft to bring Xbox PC games to GeForce NOW. The first game from this partnership, Gears 5 is now available with more set to be released in the coming months. There are now over 1,600 games on GeForce NOW, the richest content available on any cloud gaming service. Moving to Pro Visualization. Revenue of $295 million was up 31% sequentially, and down 53% year-on-year. Sequential growth was driven by stronger workstation demand across both mobile and desktop form factors, with strength in key verticals such as Public Sector, Healthcare, and Automotive. We believe the channel inventory correction is behind us. The ramp of our Ada Lovelace GPU architecture in workstations kicked-off a major product cycle. At GTC, we announced six new RTX GPUs for laptops and desktop workstations, with further rollout planned in the coming quarters. Generative AI is a major new workload for NVIDIA-powered workstation. Our collaboration with Microsoft transformed windows into the ideal platform for creators and designers, harnessing generative AI to elevate their creativity and productivity. At GTC, we announced NVIDIA Omniverse Cloud, an NVIDIA fully managed service running in Microsoft Azure that includes the full suite of Omniverse applications and NVIDIA OVX infrastructure. Using this full stack cloud environment, customers can design, develop, deploy, and manage industrial metaverse applications. NVIDIA Omniverse Cloud will be available starting in the second half of this year. Microsoft NVIDIA will also connect Office 365 applications with Omniverse. Omniverse Cloud is being used by companies to digitalize their workflows from design and engineering to smart factories and 3D content generation for marketing. The automotive industry has been a leading early adopter of Omniverse, including companies such as BMW Group, Geely Lotus, General Motors, and Jaguar Land Rover. Moving to Automotive. Revenue was $296 million, up 1% sequentially, and up 114% from a year ago. Our strong year-on-year growth was driven by the ramp of the NVIDIA DRIVE Orin across a number of new energy vehicles. As we announced in March, our automotive design win pipeline over the next six years now stands at $14 billion, up from $11 billion a year ago, giving us visibility into continued growth over the coming years. Sequentially, growth moderated as some NEV customers in China are adjusting their production schedules to reflect slower than expected demand growth. We expect this dynamic to linger for the rest of the calendar year. During the quarter, we expanded our partnership with BYD, the world's leading manufacturer of NEVs. Our new design win will extend BYD's use of the DRIVE Orin to its next-generation high-volume Dynasty, and Ocean series of vehicles, set to start production in calendar 2024. Moving to the rest of the P&L. GAAP gross margins was 64.6%, and non-GAAP gross margins were 66.8%. Gross margins have now largely recovered to prior peak level, and we have absorbed higher costs, and offset them by innovating and delivering higher valued products as well as products incorporating more and more software. Sequentially, GAAP operating expenses were down 3%, and non-GAAP operating expenses were down 1%, We have held OpEx at roughly the same level over the last past four quarters. We're working through the inventory corrections in gaming and professional visualization. We now expect to increase investments in the business while also delivering operating leverage. We returned $99 million to shareholders in the form of cash dividends. At the end of the Q1, we have approximately $7 billion remaining under our share repurchase authorization through December 2023. Let me turn to the outlook for the second quarter fiscal '24. Total revenue is expected to be $11 billion, plus or minus 2%. We expect this sequential growth to largely be driven by data center, reflecting a steep increase in demand related to generative AI and large language models. This demand has extended our data center visibility out a few quarters and we have procured substantially higher supply for the second half of the year. GAAP and non-GAAP gross margins are expected to be 68.6% and 70% respectively, plus or minus 50 basis points. GAAP and non-GAAP operating expenses are expected to be approximately $2.71 billion and $1.9 billion, respectively. GAAP and non-GAAP other income and expenses are expected to be an income of approximately $90 million, excluding gains and losses from non-affiliated investments. GAAP and non-GAAP tax rates are expected to be 14%, plus or minus 1%, excluding any discrete items. Capital expenditures are expected to be approximately $300 million to $350 million. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight some of the upcoming events, Jensen will give the COMPUTEX keynote address in person in Taipei this coming Monday, May 29 local time, which will be Sunday evening in the U.S. In addition, we will be attending the BofA Global Technology Conference in San Francisco on June 6. And Rosenblatt Virtual Technology Summit on The Age of AI on June 7, and the New Street Future of Transportation Virtual Conference on June 12. Our earnings call to discuss the results of our second quarter fiscal '24 is scheduled for Wednesday, August 23. Well, that covers our opening remarks. We're now going to open the call for questions. Operator, would you please poll for questions?

Toshiya Hari:

Hi. Good afternoon. Thank you so much for taking the question and congrats on the strong results, and incredible outlook. Just one question on data center. Colette, you mentioned the vast majority of the sequential increase in revenue this quarter will come from data center. I was curious what the construct is there, if you can speak to, what the key drivers are from April to July and perhaps more importantly, you talked about visibility into the second half of the year. I'm guessing it's more of a supply problem at this point, what kind of sequential growth beyond the July quarter can your supply chain support at this point? Thank you.

Jensen Huang:

I thought that was great color. Thank you.

C.J. Muse:

Yeah. Good afternoon. Thank you for taking the question. I guess with data center, you are essentially doubling quarter-on-quarter, two natural kind of questions that relate to one another come to mind. Number one, where are we in terms of driving acceleration into servers to support AI? And as part of that, as you deal with longer cycle times with TSMC and your other partners, how are you thinking about managing their commitments there with where you want to manage your lead times in the coming years to best match that supply and demand? Thanks so much.

Operator:

And we'll move to our next question, Vivek Arya with BofA Securities. Your line is open.

Colette Kress:

Yeah, Vivek. Thanks for the question. Let me see if I can add a little bit more color. We believe that the supply that we will have for the second half of the year will be substantially larger than H1. So, we are expecting not only the demand that we just saw in this last quarter, the demand that we have in Q2 for our forecast, but also planning on seeing something in the second half of the year. We just have to be careful here. But we are not here to guide on the second half of that. Yes, we do plan a substantial increase in the second half compared to the first half.

Operator:

Next we go to Aaron Rakers with Wells Fargo. Your line is open.

Colette Kress:

Thanks for the question. Software is really important to our accelerated platforms. Not only do we have a substantial amount of software that we are including in our nearest architecture and essentially, all products that we have. We are now with many different models to help customers start their work in generative AI and accelerated computing. So, anything that we have here from a DGX Cloud and providing those services, helping them build models or as we've discussed the importance of NVIDIA AI enterprise, essentially that operating system for AI. So, all things should continue to grow as we go forward, both the architecture and the infrastructure, as well as both availability of this offering, our ability to monetize [indiscernible] as well. I'll turn it over to Jensen, if he needs to add.

Operator:

Next we'll go to Timothy Arcuri with UBS. Your line is Open.

Jensen Huang:

Colette, how about you go first? You should take the question.

Jensen Huang:

InfiniBand and Ethernet are Target different applications in a data center. All right. They both have their place. InfiniBand had a record quarter. We're going to have a giant record year. And InfiniBand has a really -- NVIDIA's Quantum InfiniBand has an exceptional roadmap. It's going to be really incredible. But the two networks are very different. InfiniBand is designed for an AI factory, if you will. If that data center is running a few applications for a few people for a specific use case and it's doing it continuously and that infrastructure costs you, pick a number, $500 million. The difference between InfiniBand and Ethernet could be 15%, 20% in overall throughput. And if you spent $500 million in an infrastructure and the difference is 10% to 20% and it's a $100 million, InfiniBand is basically free. That's the reason why people use it. InfiniBand is effectively free. The difference in data center throughput is just -- it's too great to ignore, and you're using it for that one application and so, however, if your data center is a cloud datacenter and its multi-tenant. It's a bunch of little jobs, a bunch of little jobs and is shared by millions of people. Then Ethernet is really do I answer? There's a new segment in the middle where the Cloud is becoming a generative AI cloud. It's not only AI factory per se. But it's still a multi-tenant Cloud but it wants to run generative AI workloads. This new segment is a wonderful opportunity and at COMPUTEX, I referred to it at the last GTC. At COMPUTEX, we're going to announce a major product line for this segment, which is Ethernet focused generative AI application type of clouds. But InfiniBand is doing fantastically and we're doing record numbers quarter-on-quarter year-on-year.

Stacy Rasgon:

Hi, guys. Thanks for taking my question. I had a question on inference versus training for generative AI. So, you're talking about inference as being a very large opportunity. I guess, two sub parts to that. Is that, besides inference basically scales with like the usage versus like training is more of a one-and-done. And can you give us some sort of even if it's just like qualitatively, like if do you think are influence is bigger than training or vice-versa, like if it's bigger, how much bigger? Is it like the opportunity, is it 5x, is it 10x, is there anything you can give us on those two workloads within generative AI, it would be helpful.

Operator:

Next, we'll go to Joseph Moore with Morgan Stanley. Your line is open.

Jensen Huang:

Yeah. That's a great question. Whether your -- you start by building a large language model and you use that large language model very large version and you could distill them into medium, small and tiny size. And the tiny sized ones, you can put in your phone and your PC and so on and so forth and they all have good -- they all have -- it seems surprising, but they all can do the same thing. But obviously, the zero shot or the generalizeability of the large language model, the biggest one is much more versatile and it can do a lot more amazing things. And the large one would teach the smaller ones, how to be good AIs and so, you use the large one to generate prompts to align the smaller ones and so on and so forth. And so you start by building very large ones. And then you also have to train a whole bunch of smaller ones. Now, that's exactly the reason why we have so many different sizes of our inference. You saw that I announced L4, L40, H100 NBL -- which also have H100 HGX and then we have H100 multi-node with NVLink and so there is -- you could have model sizes of any kind that you like. The other thing that's important is, these are models, but they are connected ultimately to applications. And the applications could have image in, video out, video in, text out, image in, proteins out, text in, 3D out, video in, in the future, 3D graphics out. So, the input and the output requires a lot of pre and post-processing. The pre and post-processing can't be ignored. And this is one of the things that most of the specialized chip arguments fall apart and it's because the length -- the model itself is only call it 25% of the data -- of the overall processing of inference. The rest of it is about preprocessing and post-processing, security, decoding, all kinds of things like that. And so, I think the multimodality aspect of inference, the multi diversity of inference, that it's going to be done in the Cloud on-prem. It's going to be done in multi-cloud, that's the reason why we have the AI Enterprise in all the clouds. It's going to be done on-prem, it's the reason why we have a great partnership with Dell we just announced the other day, called project Helix, that's going to be integrated into third-party services. That's the reason why we have a great partnership with ServiceNow, and Adobe, because they're going to be creating a whole bunch of generative AI capabilities. And so, there's all the diversity, and the reach of generative AI is so broad, you need to have some very fundamental capabilities like what I just described, in order to really address the whole space of it.

Harlan Sur:

Hi. Good afternoon, and congratulations on the strong results and execution. I really appreciate more of the focus or some of the focus today in your networking products. I mean, it's really an integral part to sort of maximize the full performance of your compute platforms. I think so data center networking business is driving a part of $1 billion of revenues per quarter plus or minus, that's 2.5x growth from three years ago, right, when you guys acquired Mellanox. So very strong growth, but given the very high attach of your InfiniBand, Ethernet solutions, your accelerated compute platforms, is the networking run-rate stepping up in line with your compute shipment? And then, what is the team doing to further unlock more networking bandwidth going forward just to keep pace with the significant increase in compute complexity, datasets, requirements for lower latency, better traffic predictability, and so on?

Operator:

And our last question will come from Matt Ramsay with TD Cowen. Your line is open.

Jensen Huang:

Yeah. Thanks Matt. It's -- without being too specific about numbers, but the ideal scenario, the ideal mix is something like 10% NVIDIA DGX Cloud and 90% the CSPs clouds, and the reason -- and our DGX Cloud is -- the NVIDIA stack is the pure NVIDIA stack. It is architected the way we like and achieves the best possible performance. It gives us the ability to partner very deeply with the CSPs to create the highest-performing infrastructure, number one. Number two, it allows us to partner with the CSPs to create markets like, for example, we're partnering with Azure to bring Omniverse cloud to the world's industries. And the world's never had a system like that, the computing stack. Now with all the generative AI stuff and all the 3D stuff and the physics stuff, incredibly large database and really high-speed networks and low-latency networks, that kind of a virtual industrial virtual world has never existed before. And so, we partnered with Microsoft to create Omniverse cloud inside Azure cloud. So, it allows us number two, to create new applications together and develop new markets together. And we go-to-market as one team and we benefit by getting our customers on our computing platform and they benefit by having us in their cloud, number one; but number two, the amount of data and services and security services and all of the amazing things that Azure and GCP and OCI have, they can instantly have access to that through Omniverse cloud. And so it's a huge win-win. And for the customers, the way that NVIDIA's cloud works for these early applications, they can do it anywhere. So one standard stack runs in all the clouds and if they would like to take their software and run it on the CSPs cloud themselves and manage it themselves, we're delighted by that, because NVIDIA AI Enterprise, NVIDIA AI Foundations. And long-term, this is going to take a little longer, but NVIDIA Omniverse will run in the CSPs clouds. Okay. So, our goal really is to drive architecture to partner deeply in creating new markets and the new applications that we're doing and provide our customers with the flexibilities to run in their -- in their everywhere, including on-prem and so, that -- those were the primary reasons for it and it's worked out incredibly. Our partnership with the three CSPs and that we currently have DGX Cloud in and their sales force and marketing teams, their leadership team is really quite spectacular. It works great.

Jensen Huang:

The computer industry is going through two simultaneous transitions, accelerated computing and generative AI. CPU scaling has slowed, yet computing demand is strong and now with generative AI supercharged. Accelerated computing, a full stack and data center scale approach that NVIDIA pioneered is the best path forward. There is $1 trillion installed in the global data center infrastructure, based on the general-purpose computing method of the last era. Companies are now racing to deploy accelerated computing for the generative AI era. Over the next decade, most of the world's data centers will be accelerated. We are significantly increasing our supply to meet the surging demand. Large language models can learn information encoded in many forms. Guided by large language models, generative AI models can generate amazing content and with models to fine-tune, guardrail, align to guiding principles and ground the facts, generative AI is emerging from labs and is on its way to industrial applications. As we scale with cloud and Internet service providers, we are also building platforms for the world's largest enterprises. Whether within one of our CSP partners or on-prem with Dell Helix, whether on a leading enterprise platform like ServiceNow and Adobe or a bespoke with NVIDIA AI Foundations, we can help enterprises leverage their domain expertise and data to harness generative AI securely and safely. We are ramping a wave of products in the coming quarters, including H100, our Grace and Grace Hopper super chips and our BlueField-3 and Spectrum 4 networking platform. They are all in production. They will help deliver data center scale computing that is also energy-efficient and sustainable computing. Join us next week at COMPUTEX and we'll show you what's next. Thank you.

Operator:

Good afternoon. My name is Emma, and I will be your conference operator today. At this time, I would like to welcome everyone to the NVIDIA's Fourth Quarter Earnings Call. [Operator Instructions]. Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thank you, Simona. Q4 revenue was $6.05 billion, up 2% sequentially, while down 21% year-on-year. Full year revenue was $27 billion, flat from the prior year. Starting with data center. Revenue was $3.62 billion was down 6% sequentially and up 11% year-on-year. Fiscal year revenue was $15 billion and up 41%. Hyperscale customer revenue posted strong sequential growth, though short of our expectations as some cloud service providers paused at the end of the year to recalibrate their build plans. Though we generally see tightening that reflects overall macroeconomic uncertainty, we believe this is a timing issue as the end market demand for GPUs and AI infrastructure is strong. Networking grew but a bit less than our expected on softer demand for general purpose CPU infrastructure. The total data center sequential revenue decline was driven by lower sales in China, which was largely in line with our expectations, reflecting COVID and other domestic issues. With cloud adoption continuing to grow, we are serving an expanding list of fast-growing cloud service providers, including Oracle and GPU specialized CSPs. Revenue growth from CSP customers last year significantly outpaced that of Data Center as a whole as more enterprise customers moved to a cloud-first approach. On a trailing 4-quarter basis, CSP customers drove about 40% of our Data Center revenue. Adoption of our new flagship H100 data center GPU is strong. In just the second quarter of its ramp, H100 revenue was already much higher than that of A100, which declined sequentially. This is a testament of the exceptional performance on the H100, which is as much as 9x faster than the A100 for training and up 30x faster than [inferencing of] (ph) transformer-based large language models. The transformer engine of H100 arrived just in time to serve the development and scale out of inference of large language models. AI adoption is at an inflection point. Open AI's ChatGPT has captured interest worldwide, allowing people to experience AI firsthand and showing what's possible with generative AI. These new types of neural network models can improve productivity in a wide range of tasks, whether generating text like marketing copy, summarizing documents like [indiscernible], creating images for ads or video games or answering customer questions. Generative AI applications will help almost every industry do more faster. Generative large language models with over 100 billion parameters are the most advanced neural networks in today's world. NVIDIA's expertise spans across the AI supercomputers, algorithms, data processing and training methods that can bring these capabilities to enterprise. We look forward to helping customers with generative AI opportunities. In addition to working with every major hyperscale cloud provider, we are engaged with many consumer Internet companies, enterprises and start-ups. The opportunity is significant and driving strong growth in the data center that will accelerate through the year. During the quarter, we made notable announcements in the financial services sector, one of our largest industry verticals. We announced a partnership with Deutsche Bank to accelerate the use of AI and machine learning in financial services. Together, we are developing a range of applications, including virtual customer service agents, speech AI, fraud detection and bank process automation, leveraging NVIDIA's full computing stack, both on-premise and in the cloud, including NVIDIA AI enterprise software. We also announced that NVIDIA captured leading results for AI inference in a key financial services industry benchmark for applications such as asset price discovery. In networking, we see growing demand for our latest generation InfiniBand and HPC optimized Ethernet platforms fueled by AI. Generative AI foundation model sizes continue to grow at exponential rates, driving the need for high-performance networking to scale out multi-node accelerated workloads. Delivering unmatched performance, latency and in-network computing capabilities, InfiniBand is the clear choice for power-efficient cloud scale, generative AI. For smaller scale deployments, NVIDIA is bringing its full accelerated stack expertise and integrating it with the world's most advanced high-performance Ethernet fabrics. In the quarter, InfiniBand led our growth as our Quantum 2 40 gigabit per second platform is off to a great start, driven by demand across cloud, enterprise and supercomputing customers. In Ethernet, our 40 gigabit per second Spectrum 4 networking platform is gaining momentum as customers transition to higher speeds, next-generation adapters and switches. We remain focused on expanding our software and services. We released version 3.0 of NVIDIA AI enterprise with support for more than 50 NVIDIA AI frameworks and pretrained model and new workflows for contact center intelligent virtual assistance, audio transcription and cybersecurity. Upcoming offerings include our NeMo and BioNeMo large language model services, which are currently in early access with customers. Now to Jensen to talk a bit more about our software and cloud business.

Colette Kress:

Thanks, Jensen. Gaming revenue of $1.83 billion was up 16% sequentially and down 46% from a year ago. Fiscal year revenue of $9.07 billion is down 27%. Sequential growth was driven by the strong reception of our 40 Series GeForce RTX GPUs based on the Ada Lovelace architecture. The year-on-year decline reflects the impact of channel inventory correction, which is largely behind us. And demand in the seasonally strong fourth quarter was solid in most regions. While China was somewhat impacted by disruptions related to COVID, we are encouraged by the early signs of recovery in that market. Gamers are responding enthusiastically to the new RTX4090, 4080, 4070 Ti desktop GPUs, with many retail and online outlets quickly selling out of stock. The flagship RTX 4090 has quickly shot up in popularity on Steam to claim the top spot for the AI architecture, reflecting gamers' desire for high-performance graphics. Earlier this month, the first phase of gaming laptops based on the Ada architecture reached retail shelves, delivering NVIDIA's largest-ever generational leap in performance and power efficiency. For the first time, we are bringing enthusiast-class GPU performance to laptops as slim as 14 inches, a fast-growing segment, previously limited to basic tasks and apps. In another first, we are bringing the 90 class GPUs, our most performing models, to laptops, thanks to the power efficiency of our fifth-generation Max-Q technology. All in, RTX 40 Series GPUs will power over [170] (ph) gaming and creator laptops, setting up for a great back-to-schools season. There are now over 400 games and applications supporting NVIDIA's RTX technology for real-time ray tracing and AI-powered graphics. The AI architecture features DLSS 3, our third-generation AI-powered graphics, which massively boosts performance. With the most advanced games, Cyberpunk 2077, recently added DLSS 3 enabling a 3 to 4x boost in frame rate performance at 4K resolution. Our GeForce NOW cloud gaming service continued to expand in multiple dimensions, users, titles and performance. It now has more than 25 million members in over 100 countries. Last month, it enabled RTX 4080 graphics horsepower in the new high-performance ultimate membership tier. Ultimate members can stream at up to 240 frames per second from a cloud with full ray tracing and DLSS 3. And just yesterday, we made an important announcement with Microsoft. We agreed to a 10-year partnership to bring to GeForce NOW Microsoft's lineup of Xbox PC games, which includes blockbusters like Minecraft, Halo and Flight Simulator. And upon the close of Microsoft's Activision acquisition, it will add titles like Call of Duty and Overwatch. Moving to Pro Visualization. Revenue of $226 million was up 13% sequentially and down 65% from a year ago. Fiscal year revenue of $1.54 billion was down 27%. Sequential growth was driven by desktop workstations with strengths in the automotive and manufacturing industrial verticals. Year-on-year decline reflects the impact of the channel inventory correction, which we expect to end in the first half of the year. Interest in NVIDIA's Omniverse continues to build with almost 300,000 downloads so far, 185 connectors to third-party design applications. The latest released Omniverse has a number of features and enhancements, including support for 4K, real-time path tracing, Omniverse Search for AI-powered search through large untagged 3D databases, and Omniverse cloud containers for AWS. Let's move to automotive. Revenue was a record $294 million, up 17% from [indiscernible] and up 135% from a year ago. Sequential growth was driven primarily by AI automotive solutions. New program ramps at both electric vehicle and traditional OEM customers helped drive this growth. Fiscal year revenue of $903 million was up 60%. At CES, we announced a strategic partnership with Foxconn to develop automated and autonomous vehicle platforms. This partnership will provide scale for volume, manufacturing to meet growing demand for the NVIDIA Drive platform. Foxconn will use NVIDIA Drive, Hyperion compute and sensor architecture for its electric vehicles. Foxconn will be a Tier 1 manufacturer producing electronic control units based on NVIDIA Drive Orin for the global . We also reached an important milestone this quarter. The NVIDIA Drive operating system received safety certification from TÜV SÜD, one of the most experienced and rigorous assessment bodies in the automotive industry. With industry-leading performance and functional safety, our platform meets the higher standards required for autonomous transportation. Moving to the rest of the P&L. GAAP gross margin was 63.3%, and non-GAAP gross margin was 66.1%. Fiscal year GAAP gross margin was 56.9%, and non-GAAP gross margin was 59.2%. Year-on-year, Q4 GAAP operating expenses were up 21%, and non-GAAP operating expenses were up 23%, primarily due to the higher compensation and data center infrastructure expenses. Sequentially, GAAP operating expenses were flat, and non-GAAP operating expenses were down 1%. We plan to keep them relatively flat at this level over the coming quarters. Full year GAAP operating expenses were up 50%, and non-GAAP operating expenses were up 31%. We returned $1.15 billion to shareholders in the form of share repurchases and cash dividends. At the end of Q4, we had approximately %7 billion remaining under our share repurchase authorization through December 2023. Let me look to the outlook for the first quarter of fiscal '24. We expect sequential growth to be driven by each of our 4 major market platforms led by strong growth in data center and gaming. Revenue is expected to be $6.5 billion, plus or minus 2%. GAAP and non-GAAP gross margins are expected to be 64.1% and 66.5%, respectively, plus or minus 50 basis points. GAAP operating expenses are expected to be approximately $2.53 billion. Non-GAAP operating expenses are expected to be approximately $1.78 billion. GAAP and non-GAAP other income and expenses are expected to be an income of approximately $50 million, excluding gains and losses of nonaffiliated divestments. GAAP and non-GAAP tax rates are expected to be 13%, plus or minus 1%, excluding any discrete items. Capital expenditures are expected to be approximately $350 million to $400 million for the first quarter and in the range of $1.1 billion to $1.3 billion for the full fiscal year 2024. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight upcoming events for the financial community. We will be attending the Morgan Stanley Technology Conference on March 6 in San Francisco and the Cowen Healthcare Conference on March 7 in Boston. We will also host GTC virtually with Jensen's keynote kicking off on March 21. Our earnings call to discuss the results of our first quarter of fiscal year '24 is scheduled for Wednesday, May 24. Now we will open up the call for questions. Operator, would you please poll for questions?

Aaron Rakers:

Clearly, on this call, a key focal point is going to be the monetization effect of your software and cloud strategy. I think as we look at it, I think, straight up, the enterprise AI software suite, I think, is priced at around $6,000 per CPU socket. I think you've got pricing metrics a little bit higher for the cloud consumption model. I'm just curious, Colette, how do we start to think about that monetization contribution to the company's business model over the next couple of quarters relative to, I think, in the past, you've talked like a couple of hundred million or so? Just curious if you can unpack that a little bit.

Jensen Huang:

Yes, first of all, taking a step back, NVIDIA AI is essentially the operating system of AI systems today. It starts from data processing to learning, training, to validations, to inference. And so this body of software is completely accelerated. It runs in every cloud. It runs on-prem. And it supports every framework, every model that we know of, and it's accelerated everywhere. By using NVIDIA AI, your entire machine learning operations is more efficient, and it is more cost effective. You save money by using accelerated software. Our announcement today of putting NVIDIA's infrastructure and have it be hosted from within the world's leading cloud service providers accelerates the enterprise's ability to utilize NVIDIA AI enterprise. It accelerates people's adoption of this machine learning pipeline, which is not for the faint of heart. It is a very extensive body of software. It is not deployed in enterprises broadly, but we believe that by hosting everything in the cloud, from the infrastructure through the operating system software, all the way through pretrained models, we can accelerate the adoption of generative AI in enterprises. And so we're excited about this new extended part of our business model. We really believe that it will accelerate the adoption of software.

Vivek Arya:

Just wanted to clarify, Colette, if you meant data center could grow on a year-on-year basis also in Q1? And then Jensen, my main question kind of relate to 2 small related ones. The computing intensity for generative AI, if it is very high, does it limit the market size to just a handful of hyperscalers? And on the other extreme, if the market gets very large, then doesn't it attract more competition for NVIDIA from cloud ASICs or other accelerator options that are out there in the market?

Jensen Huang:

Large language models are called large because they are quite large. However, remember that we've accelerated and advanced AI processing by a million x over the last decade. Moore's Law, in its best days, would have delivered 100x in a decade. By coming up with new processors, new systems, new interconnects, new frameworks and algorithms and working with data scientists, AI researchers on new models, across that entire span, we've made large language model processing a million times faster, a million times faster. What would have taken a couple of months in the beginning, now it happens in about 10 days. And of course, you still need a large infrastructure. And even the large infrastructure, we're introducing Hopper, which, with its transformer engine, it's new NVLink switches and its new InfiniBand 400 gigabits per second data rates, we're able to take another leap in the processing of large language models. And so I think the -- by putting NVIDIA's DGX supercomputers into the cloud with NVIDIA DGX cloud, we're going to democratize the access of this infrastructure, and with accelerated training capabilities, really make this technology and this capability quite accessible. So that's one thought. The second is the number of large language models or foundation models that have to be developed is quite large. Different countries with different cultures and its body of knowledge are different. Different fields, different domains, whether it's imaging or its biology or its physics, each one of them need their own domain of foundation models. With large language models, of course, we now have a prior that could be used to accelerate the development of all these other fields, which is really quite exciting. The other thing to remember is that the number of companies in the world have their own proprietary data. The most valuable data in the world are proprietary. And they belong to the company. It's inside their company. It will never leave the company. And that body of data will also be harnessed to train new AI models for the very first time. And so we -- our strategy and our goal is to put the DGX infrastructure in the cloud so that we can make this capability available to every enterprise, every company in the world who would like to create proprietary data and so -- proprietary models. The second thing about competition. We've had competition for a long time. Our approach, our computing architecture, as you know, is quite different on several dimensions. Number one, it is universal, meaning you could use it for training, you can use it for inference, you can use it for models of all different types. It supports every framework. It supports every cloud. It's everywhere. It's cloud to private cloud, cloud to on-prem. It's all the way out to the edge. It could be an autonomous system. This one architecture allows developers to develop their AI models and deploy it everywhere. The second very large idea is that no AI in itself is an application. There's a preprocessing part of it and a post-processing part of it to turn it into an application or service. Most people don't talk about the pre and post processing because it's maybe not as sexy and not as interesting. However, it turns out that preprocessing and post-processing oftentimes consumes half or 2/3 of the overall workload. And so by accelerating the entire end-to-end pipeline, from preprocessing, from data ingestion, data processing, all the way to the preprocessing all the way to post processing, we're able to accelerate the entire pipeline versus just accelerating half of the pipeline. The limit to speed up, even if you're instantly passed if you only accelerate half of the workload, is twice as fast. Whereas if you accelerate the entire workload, you could accelerate the workload maybe 10, 20, 50x faster, which is the reason why when you hear about NVIDIA accelerating applications, you routinely hear 10x, 20x, 50x speed up. And the reason for that is because we accelerate things end to end, not just the deep learning part of it, but using CUDA to accelerate everything from end to end. And so I think the universality of our computing -- accelerated computing platform, the fact that we're in every cloud, the fact that we're from cloud to edge, makes our architecture really quite accessible and very differentiated in this way. And most importantly, to all the service providers, because of the utilization is so high, because you can use it to accelerate the end-to-end workload and get such a good throughput, our architecture is the lowest operating cost. It's not -- the comparison is not even close. So -- anyhow those are the 2 answers.

Christopher Muse:

I guess, Jensen, you talked about ChatGPT as an inflection point kind of like the iPhone. And so curious, part A, how have your conversations evolved post ChatGPT with hyperscale and large-scale enterprises? And then secondly, as you think about Hopper with the transformative engine and Grace with high-bandwidth memory, how have you kind of your outlook for growth for those 2 product cycles evolved in the last few months?

Operator:

Your next question comes from the line of Matt Ramsay with Cowen & Company.

Jensen Huang:

Yes, I appreciate the question. First of all, as you know, our Data Center business is a GPU business only in the context of a conceptual GPU because what we actually sell to the cloud service providers is a panel, a fairly large computing panel of 8 Hoppers or 8 Amperes that's connected with NVLink switches that are connected with NVLink. And so this board represents essentially 1 GPU. It's 8 chips connected together into 1 GPU with a very high-speed chip-to-chip interconnect. And so we've been working on, if you will, multi-die computers for quite some time. And that is 1 GPU. So when we think about a GPU, we actually think about an HGX GPU, and that's 8 GPUs. We're going to continue to do that. And the thing that the cloud service providers are really excited about is by hosting our infrastructure for NVIDIA to offer because we have so many companies that we work directly with. We're working directly with 10,000 AI start-ups around the world, with enterprises in every industry. And all of those relationships today would really love to be able to deploy both into the cloud at least or into the cloud and on-prem and oftentimes multi-cloud. And so by having NVIDIA DGX and NVIDIA's infrastructure are full stack in their cloud, we're effectively attracting customers to the CSPs. This is a very, very exciting model for them. And they welcomed us with open arms. And we're going to be the best AI salespeople for the world's clouds. And for the customers, they now have an instantaneous infrastructure that is the most advanced. They have a team of people who are extremely good from the infrastructure to the acceleration software, the NVIDIA AI open operating system, all the way up to AI models. Within 1 entity, they have access to expertise across that entire span. And so this is a great model for customers. It's a great model for CSPs. And it's a great model for us. It lets us really run like the wind. As much as we will continue and continue to advance DGX AI supercomputers, it does take time to build AI supercomputers on-prem. It's hard no matter how you look at it. It takes time no matter how you look at it. And so now we have the ability to really prefetch a lot of that and get customers up and running as fast as possible.

Timothy Arcuri:

Jensen, I had a question about what this all does to your TAM. Most of the focus right now is on text, but obviously, there are companies doing a lot of training on video and music. They're working on models there. And it seems like somebody who's training these big models has maybe, on the high end, at least 10,000 GPUs in the cloud that they've contracted and maybe tens of thousands of more to inference a widely deployed model. So it seems like the incremental TAM is easily in the several hundred thousands of GPUs and easily in the tens of billions of dollars. But I'm kind of wondering what this does to the TAM numbers you gave last year. I think you said $300 billion hardware TAM and $300 billion software TAM. So how do you kind of think about what the new TAM would be?

Operator:

Your next question comes from the line of Stacy Rasgon with Bernstein.

Colette Kress:

Great. Thanks, Stacy, for the question. Let me first start with your question you had about H-100 and A100. We began initial shipments of H-100 back in Q3. It was a great start. Many of them began that process many quarters ago. And this was a time for us to get production level to them in Q3. So Q4 was an important time for us to see a great ramp of H-100 that we saw. What that means is our H-100 was the focus of many of our CSPs within Q4, and they were all wanting to get both get up and running in cloud instances. And so we actually saw less of A100 in Q4 of what we saw in H-100 at a larger amount. We tend to continue to sell both architectures going forward, but just in Q4, it was a strong quarter for Your additional questions that you had on Mercedes Benz. I'm very pleased with the joint connection that we have with them and the work. We've been working very diligently about getting ready to come to market. And you're right. They did talk about the software opportunity. They talked about their software opportunity in 2 phases, about what they can do with Drive as well as what they can also do with Connect. They extended out to a position of probably about 10 years looking at the opportunity that they see in front of us. So it aligns with what our thoughts are with a long-term partner of that and sharing that revenue over time.

Operator:

Your next question comes from the line of Mark Lipacis with Jefferies.

Jensen Huang:

Yes, Mark, I really appreciate the question. First of all, I have new applications that you don't know about and new workloads that we've never shared that I would like to share with you at GTC. And so that's my hook to come to GTC, and I think you're going to be very surprised and quite delighted by the applications that we're going to talk about. Now there's a reason why it is the case that you're constantly hearing about new applications. The reason for that is, number one, NVIDIA is a multi-domain accelerated computing platform. It is not completely general purpose like a CPU because a CPU is 95%, 98% control functions and only 2% mathematics, which makes it completely flexible. We're not that way. We're an accelerated computing platform that works with the CPU that offloads the really heavy computing units, things that could be highly, highly paralyzed to offload them. But we're multi-domain. We could do particle systems. We could do fluids. We could do neurons. And we can do computer graphics. We can do . There are all kinds of different applications that we can accelerate, number one. Number two, our installed base is so large. This is the only accelerated computing platform, the only platform. Literally, the only one that is architecturally compatible across every single cloud from PCs to workstations, gamers to cars to on-prem. Every single computer is architecturally compatible, which means that a developer who developed something special would seek out our platform because they like the reach. They like the universal reach. They like the acceleration, number one. They like the ecosystem of programming tools and the ease of using it and the fact that they have so many people they can reach out to, to help them. There are millions of CUDA experts around the world, software all accelerated, tool all accelerated. And then very importantly, they like the reach. They like the fact that you can see -- they can reach so many users after they develop the software. And it is the reason why we just keep attracting new applications. And then finally, this is a very important point. Remember, the rate of CPU computing advance has slowed tremendously. And whereas back in the first 30 years of my career, at 10x in performance at about the same power every 5 years and then 10x every 5 years. That rate of continued advance has slowed. At a time when people still have really, really urging applications that they would like to bring to the world, and they can't afford to do that with the power keep going up. Everybody needs to be sustainable. You can't continue to consume power. By accelerating it, we can decrease the amount of power you use for any workload. And so all of these multitude of reasons is really driving people to use accelerated computing, and we keep discovering new exciting applications.

Atif Malik:

Colette, I have a question on data center. You saw some weakness on build plan in the January quarter, but you're guiding to year-over-year acceleration in April and through the year. So if you can just rank order for us the confidence in the acceleration. Is that based on your H-100 ramp or generative AI sales coming through or the new AI services model? And also, if you can talk about what you're seeing on the enterprise vertical.

Jensen Huang:

No, you did great. That was great.

Joseph Moore:

Jensen, you talked about the sort of 1 million times improvement in your ability to train these models over the last decade. Can you give us some insight into what that looks like in the next few years and to the extent that some of your customers with these large language models are talking about 100x the complexity over that kind of time frame. I know Hopper is 6x better transformer performance. But what can you do to scale that up? And how much of that just reflects that it's going to be a much larger hardware expense down the road?

Operator:

This concludes our Q&A session. I will now turn the call back over to Jensen Huang for closing remarks.

Operator:

This concludes today's conference. You may now disconnect.

Operator:

Good afternoon. My name is Emma, and I will be your conference operator today. At this time, I would like to welcome everyone to the NVIDIA's third quarter earnings call. [Operator instructions] Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. Q3 revenue was $5.93 billion, down 12% sequentially and down 17% year on year. We delivered record data center and automotive revenue. while our gaming and pro visualization platforms declined as we work through channel inventory corrections and challenging external conditions. Starting with data center. Revenue of $3.83 billion was up 1% sequentially and 31% year-on-year. This reflects very solid performance in the face of macroeconomic challenges new export controls and lingering supply chain disruptions. Year-on-year growth was driven primarily by leading U.S. cloud providers and a broadening set of consumer Internet companies for workloads such as large language models, recommendation systems and generative AI. As the number and scale of public cloud computing and Internet service companies deploying NVIDIA AI grows our traditional hyperscale definition will need to be expanded to convey the different end market use cases. We will align our data center customer commentary going forward accordingly. Other vertical industries, such as automotive and energy, also contributed to growth with key workloads relating to autonomous driving, high-performance computing, simulations and analytics. During the quarter, the U.S. government announced new restrictions impacting exports of our A100 and H-100 based products to China, and any product destined for certain systems or entities in China. These restrictions impacted third quarter revenue, largely offset by sales of alternative products into China. That said, demand in China more broadly remains soft, and we expect that to continue in the current quarter. We started shipping our flagship 100 data center GPU based on the new hopper architecture in Q3. A100-based systems are available starting this month from leading server makers including Dell, Hewlett Packard Enterprise, Lenovo and SuperMicro. Early next year, the first H-100 based cloud instances will be available on Amazon Web Services, Google Cloud, Microsoft Azure and Oracle Cloud Infrastructure. A100 delivered the highest performance and workload versatility for both AI training and inference in the latest MLPerf industry benchmarks. H-100 also delivers incredible value compared to the previous generation for equivalent AI performance it offers three x lower total cost of ownership while using five x fewer server nodes and 3.5 x less energy. Earlier today, we announced a multiyear collaboration with Microsoft to build an advanced cloud-based AI supercomputer to help enterprises train, deploy and scale AI including large state-of-the-art models. MacBook Azure will incorporate our complete AI stack, adding tens and thousands of A100 and A100 GPUs. Quantum 2 400 gigabit per second InfiniBand networking and the NVIDIA AI enterprise software suite to its platform. Oracle and NVIDIA are also working together to offer AI training and inference at scale to thousands of enterprises. This includes bringing to Oracle Cloud infrastructure, the full NVIDIA accelerated computing stack and adding tens of thousands of NVIDIA GPUs, including the A100 and H-100. Cloud-based high-performance in the company, new scale is adopting NVIDIA AI enterprise and other software to address the industrial scientific communities, rising demand for AI in the cloud. NVIDIA AI will bring new capability to rescale high-performance computing as a service offerings, which include simulation and engineering software used across industries. Networking posted strong growth driven by hyperscale customers and easing supply constraints. -- our new Quantum 240 gigabit per second InfiniBand and Spectrum Ethernet networking platforms are building momentum. We achieved an important milestone this quarter with VMware. And whose leading server virtualization platform, vSphere, has been rearchitected over the last two years to run on DPUs and now supports our BlueField DPUs. Our joint enterprise AI platform is available first on Dell PowerEdge servers. The BlueField DPU design win pipeline is growing and the number of infrastructure softer partners is expanding, including Arista, Check Point, Juniper, [Inaudible] Networks and Red Hot. The latest top 500 list of supercomputers released this week at Supercomputing '22 and has the highest ever number of NVIDIA-powered systems, including 72% of the total and 90% of new systems on the list. Moreover, NVIDIA powers 23 of the top 30 of the Green 500 list, demonstrating the energy efficiency of accelerated computing. The No. 1 most energy-efficient system is the Flat Iron Institute Henry, which is the first top 500 system featuring our H-100 GPUs. At GTC, we announced the NVIDIA Omniverse Computing System, or OVS, reference designs featuring the new L4 GPU based on the ADA Lovelace architecture. These systems are designed to build and operate 3D virtual world using NVIDIA Omniverse enterprise. NVIDIA OBX systems will be available from Inspur, Lenovo and Super Micro by early 2023. We Lockheed Martin and Jaguar Land Rover will be among the first customers to receive OVS systems. We are further expanding our AI software and services offerings with NVIDIA and Bio Nemo large language model services, which are both entering early access this month. These enable developers to easily adopt large language models and deploy customized AI applications for content generation, tech summarization, chatbox, co-development, protein structure and biomolecular property predictions. Moving to gaming. Revenue of $1.57 billion was down 23% sequentially and down 51% from a year ago, reflecting lower sell-in to partners to help align channel inventory levels with current demand expectations. We believe Channel inventories are on track to approach normal levels as we exit Q4. Sell-through for our gaming products was relatively solid in the Americas and EMEA and but softer in Asia Pac as macroeconomic conditions and covered lockdowns in China continued to weigh on consumer demand. Our new Ada Lovelace GPU architecture had an exceptional launch. The first ADA GPU, the GeForce RTX 4090 became available in mid-October and a tremendous amount and positive feedback from the gaming community. We sold out quickly in many locations and are working hard to keep up with demand. The next member of the ATA family, RTX 4080 is available today. The RTX 40 Series GPUs features DLSS 3, the neuro rendering technology that uses AI to generate entire frames for faster game play. Our third-generation RTX technology has raised the bar for computer graphics and help supercharge gaming. For example, the 15-year old classic game portal, now reimagined with full ray tracing and DLSS 3 has made it on Steam's top 100 most wish-listed gains. The total number of RTX games and applications now exceeds 350. There is tremendous energy in the gaming community that we believe will continue to fuel strong fundamentals over the long term. The number of simultaneous users on steam just hit a record of $30 million, surpassing the prior peak of $28 million in January. Activision's Call of Duty Modern Warfare 2 set a record for the franchise with more than $800 million in opening weekend sales. topping the combined box office openings of movie blockbusters, TopGun Maverick and Dr. Strains in the Multiverse of [Inaudible]. And this month's League of Legends World Championship in San Francisco sold out minutes with 18,000 esports fans packed the arena where the Golden State Warriors play. We continue to expand the GeForce NOW cloud gaming service. In Q3, we added over 85 games to the library, bringing the total to over 1,400. We also launched GeForce now on the new gaming devices, including Logitech, Cloud handheld, cloud gaming Chromebooks and Razor 5G Edge. Moving to Probi Revenue of $200 million was down 60% sequentially and down 65% from a year ago, reflecting lower sell-in to partners to help align channel inventory levels with the current demand expectations. These dynamics are expected to continue in Q4. Despite near-term challenges, we believe our long-term opportunity remains intact, fueled by AI simulation, computationally intensive design and engineering workloads. At GTC, we announced NVIDIA Omniverse Cloud Services, our first software and infrastructure as a service offering, enabling artists, developers and enterprise teams to design, publish and operate metaverse applications from anywhere on any device. Omniverse Cloud Services runs on Omniverse cloud computer, a computing system comprised of NVIDIA OBX for graphics and physics simulation. NVIDIA HDX for AI workloads and the NVIDIA graphics delivery network, a global scale, distributed data center network for delivering low-latency metaverse graphics on the edge. Leaders in some of the world's largest industries continue to adopt Omniverse. Home improvement retailer, Lowe's is using it to help design, build and operate digital twins for their stores. Charter Communications and advanced analytics company, heavy AI are creating Omniverse power digital twins to optimize Charter's wireless network. In Deutsche Bahn, operator of German National Railway is using Omniverse to create digital twins of its rail network and train AI models to monitor the network, increasing safety and reliability. Moving to automotive. Revenue of $251 million, increased 14% sequentially and 86% from a year ago. Growth was driven by an increase in AI automotive solutions as our customers drive or on-based production ramp, continue to scale. Automotive has great momentum and is on its way to be our next multibillion-dollar platform. Global cars unveiled the all-new flagship Volvo EX90 SUV powered by the NVIDIA Drive platform. This is the first model to use Volvo's software-defined architecture with a centralized core computer containing both drive Orin and DRIVEXaviar, along with 30 sensors. Other recently announced design wins and new model introductions include ton, auto, Neo, Polystar and [Inaudible]. At GTC, we also announced that NVIDIA Drive Super Chip, the successor to Orin in our automotive SoC road map, drive [Inaudible] delivers up to 2,000 tariff lots of performance and leverages technologies introduced in our Grace Hopper and ADA architectures. It is capable of running both the automated drive and in-vehicle infotainment systems. Simultaneously offering a LIFA performance while reducing cost and energy consumption. Driver will be available for automakers 25 models with Geely owned automaker, Zika as the first announced customer. Moving to the rest of the P&L. GAAP gross margin was 53.6% and and non-GAAP gross margin was 56.1%. Gross margins reflect $702 million in inventory charges largely related to lower data center demand in China, partially offset by a warranty benefit of approximately $70 million. Year-on-year, GAAP operating expenses were up 31%, and non-GAAP operating expenses were up 30%, primarily due to higher compensation expenses related to headcount growth and salary increases and higher data center infrastructure expenses. Sequentially, both GAAP and non-GAAP operating expense growth was in the single-digit percent, and we plan to keep it relatively flat at these levels over the coming quarters. We returned $3.75 billion to shareholders in the form of share repurchases and cash dividends. At the end of Q3, we had approximately $8.3 billion remaining under our share repurchase authorization through December 23. Let me turn to the outlook for the fourth quarter of fiscal 2023. We expect our data center revenue to reflect early production shipments of the A100, offset by continued softness in China. In gaming, we expect to resume sequential growth with our revenue still below end demand as we continue to work through the channel inventory correction. And in automotive, we expect the continued ramp of our Oren design wins. All in, we expect modest sequential growth driven by automotive, gaming and data center. Revenue is expected to be $6 billion, plus or minus 2%. GAAP and non-GAAP gross margins are expected to be $63.2 million and 66%, respectively, plus or minus 50 basis points. GAAP operating expenses are expected to be approximately $2.56 billion. Non-GAAP operating expenses are expected to be approximately $1.78 billion. GAAP and non-GAAP other income and expenses are expected to be an income of approximately $40 million, excluding gains and losses on nonaffiliated investments. GAAP and non-GAAP tax rates are expected to be 9%, plus or minus 1%, excluding any discrete items. Capital expenditures are expected to be approximately $500 million to $550 million. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight upcoming events for the financial community. We'll be attending the Credit Suisse conference in Phoenix on November 30. The rate Virtual Tech Conference on December 5 and and the JPMorgan Forum on January 5 in Las Vegas. Our earnings call to discuss the results of our fourth quarter and fiscal 2023 are scheduled for Wednesday, February 22. We will now open the call for questions. Operator, could you please poll for questions?

Vivek Arya:

Thanks for taking my question. Colette, just wanted to clarify first, I think last quarter, you gave us a sell-through rate for your gaming business at about $2.5 billion a quarter. I think you said China is somewhat weaker. So I was hoping you could update us on what that sell-through rate is right now for gaming. And then, Jen-Hsun, the question for you. A lot of concerns about large hyperscalers cutting their spending and pointing to a slowdown. So if, let's say, U.S. cloud capex is flat or slightly down next year, do you think your business can still grow in the data center and why?

Jen-Hsun Huang:

Vivek, our data center business is indexed to two fundamental dynamics. The first has to do with general purpose computing no longer scaling. And so, acceleration is necessary to achieve the necessary level of cost efficiency scale and energy efficiency scale so that we can continue to increase workloads while saving money and saving power. Accelerated computing is recognized generally as the path forward as general purpose computing slows. The second dynamic is AI. And we're seeing surging demand in some very important sectors of AIs in important breakthroughs in AI. One is called deep recommender systems, which is quite essential now to the best content or item or product to recommend to somebody who's using a device that is like a selfie or interacting with a computer just using voice. You need to really understand the nature, the context of the person making the request and make the appropriate recommendation to them. The second has to do with large language models. This is -- this started several years ago with the invention of the transformer, which led to Bert, which led to GP3, which led to a whole bunch of other models now associated with that. We now have the ability to learn representations of languages of all kinds. It could be human language. It could be the language of biology. It could be the language of chemistry. And recently, I just saw a breakthrough called Jeans LM, we just one of the first example of learning the language of human genomes. The third has to do with generative AI. You know that the first 10 years, we've dedicated ourselves to perception AI. But the goal of perception, of course, is to understand context. But the ultimate goal of AI is to make a contribution to create something to generate product. And this is now the beginning of the era of generative AI. You probably see it all over the place, whether they're generating images or generating videos or generating text of all kinds and the ability to augment our performance to enhance our performance to make productivity enhanced to reduce cost and improve whatever we do with whatever we have to work with, productivity is really more important than ever. And so, you could see that our company is indexed to two things, both of which are more important than ever, which is power efficiency, cost efficiency and then, of course, productivity. And these things are more important than ever. And my expectation is that we're seeing all the strong demand and surging demand for AI and for niche reasons.

C.J. Muse:

Yeah, Good afternoon and thank you for taking the question. You started to bundle on NVIDIA enterprise now with the H-100. I'm curious if you can talk about how we should think about timing around software monetization? And how we should kind of see this flow through the model, particularly with the focus on the AI enterprise and Omnivere side of things?

Operator:

Your next question comes from the line of Chris Caso with Credit Suisse. Your line is now open.

Colette Kress:

Thanks for the question, Chris. So as we highlighted in our prepared remarks, we booked an entry of $702 million for inventory reserves within the quarter. Most of that, primarily, all of it is related to our data center business, just due to the change in expected demand looking forward for China. So when we look at the data center products, a good portion of this was also the A100, which we wrote down. Now, looking at our inventory that we have on hand and the inventory that has increased, a lot of that is just due to our upcoming architectures coming to market. our ADA architecture, our hopper architecture and even more in terms of our networking business. We have been building for those architectures to come to market and as such to say. We are always looking at our inventory levels at the end of each quarter for our expected demand going forward. But I think we've done a solid job that we used in this quarter just based on that expectation going forward.

Timothy Arcuri:

Thanks a lot. Colette, can you -- I have a two-part question. First, is there any effect of stockpiling in the data center guidance? I ask because you now have the A800 that is sort of a modified version of the A100 with the lower data transfer rate. So one could imagine that customers might be stocking that while they can still get it. And I guess the second part of that is related to the inventory charge, can you just go into that a little bit more? Because last quarter, it made sense that you took a charge because revenue was less than you thought, but revenue came in pretty much in line. And it sounded like China was a net neutral. So is the charge related to just working A100 inventory down faster? Is that what the charges related to?

Operator:

Your next question comes from the line of Stacy Rasgon with Bernstein. Your line is now open.

Colette Kress:

Yes. So your question is regarding the sequentials from Q3 to our guidance that we provided for Q4. As we are seeing the numbers in terms of our guidance, you're correct, is only growing about $100 million. And we've indicated that three of those platforms will likely grow just a little bit. But our pro visualization business we think is going to be flattish and likely not growing as we're still working on correcting the channel inventory levels. to get to the right amount. It's very difficult to say which will have that increase. But again, we are planning for all three of those different market platforms to grow just a little bit.

Mark Lipacis:

Hi. Thanks for taking my question. Jen-Hsun, I think for you, you've articulated a vision for the data center we're a solution with an integrated solution set of a CPU, GPU and DPU is deployed for all workloads or most workloads, I think. Could you just give us a sense of or talk about where is this vision in the penetration cycle? And maybe talk about Grace Grace's importance for realizing that vision, what will Grace deliver versus an off-the-shelf x86 where -- do you have a sense of where Grace will get embraced first or the fastest within that vision?

Operator:

Your next question comes from the line of Harlan Sur with J.P. Morgan. Your line is now open.

Jen-Hsun Huang:

Yes. If I could take that. First, thanks for your question. Our networking, as you know, is heavily indexed to high-performance computing. We're not -- we don't serve the vast majority of commodity networking. All of our network solutions are very high end, and they're designed for data centers that move a lot of data. Now, if you have a hyperscale data center these days, and you are deploying a large number of AI applications. It is very likely that the network bandwidth that you provision has a substantial implication on the overall throughput of your data center. So the small incremental investment they make in high-performance networking translates to billions of dollars of savings slightly in provisioning the service or billions of dollars more throughput, which increases their economics. And so, these days, with disaggregated and I application, AI provisioning and data centers, high-performance networking is really quite fantastic and it pays for itself right away. But that's where we are focused in high-performance networking and provisioning AI services in -- well, the AI applications that we focus on. You might have noticed that NVIDIA and Microsoft are building one of the largest AI infrastructures in the world. And it is completely powered by NVIDIA's InfiniBand 400 gigabits per second network. And the reason for that is because that network pays for itself instantaneously. The investment that you're going to put into the infrastructure is so significant that if you were to be dragged by slow networks, obviously, the efficiency of the overall infrastructure is not as high. And so, in the places where we focus networking is really quite important. It goes all the way back to when we first announced the acquisition of Mellanox. I think at the time, they were doing about a few hundred million dollars a quarter, about $400 million a quarter. And now we're doing what they used to do in the old days, in a year, practically coming up in a quarter. And so, that kind of tells you about the growth of high-performance networking. It is an indexed to overall enterprise and data center spend but it is highly indexed to AI adoption.

Aaron Rakers:

Thanks for taking the question. I want to expand on the networking question a little bit further. When we look at the Microsoft announcement today, we think about what Meda is doing on the AI footprint that they're deploying. Jen-Hsun, can you help us understand like where your InfiniBand networking sits relative to like traditional data center switching? And maybe kind of build on that, how you're positioning spectrum for in the market, does that compete against a broader set of opportunities in the Ethernet world for AI fabric networking?

Operator:

Your next question comes from the line of Ambrish Srivastava with BMO. Your line is now open.

Colette Kress:

Thank you for the question. So looking at our compute for the quarter is about flattish. Yes, we're seeing also growth growth in terms of our networking, but you should look at our Q3 compute is about flatters with last quarter.

Operator:

Your next question comes from the line of William Stein with Truist Securities. Your line is now open.

Jen-Hsun Huang:

H-100 ramp is different than the A100 ramp in several ways. The first is that the TCO, the cost benefits, the operational cost benefits because of the energy savings because every data center is now Power Limited. And because of this incredible transformer engine that's designed for the latest AI models. The performance over Ampere is so significant that I -- and because of the pent-up demand for hopper because of these new models that are that I spoke about earlier, deep recommender systems and large language models and generative AI models. Customers are clamoring to ramp hopper as quickly as possible, and we are trying to do the same. We are all hands on deck to help the cloud service providers stand up the supercomputers. Remember, I is the only company in the world that produces and ships semi-custom supercomputers in high volume. It's a miracle to ship one supercomputer every three years. it's unheard of to ship supercomputers to every cloud service provider in a quarter. And so, we're working hand with every one of them, and every one of them are racing to stand up hoppers. We expect them to have hopper cloud services stood up in Q1. And so, we are expecting to ship some volume, we're expecting to ship production in Q4, and then we're expecting to ship large volumes in Q1. That's a faster transition than MPIR. And so, it's because of the dynamics that I described.

Matt Ramsay:

Yeah. Thank you very much. Good afternoon. I guess, Colette, I heard in your script that you had talked about maybe a new way of commenting on or reporting hyperscaler revenue in your data center business. And I wonder if you could maybe give us a little bit more detail about what you're thinking there and what sort of drove the decision? And I guess the derivative of that, Jen-Hsun, how -- that decision to talk about the data center business to hyperscalers differently. I mean, what does that mean for the business that is just a reflection of where demand is and you're going to break things out differently? Or is something changing about the mix of I guess, internal properties versus vertical industry demand within the hyperscale customer base.

Jen-Hsun Huang:

Yes. Let me double click on what Colette just said, which is absolutely right. There are two major dynamics that's happening. First, the adoption of NVIDIA in Internet service companies around the world, the number and the scale by which they're doing it has grown a lot. Internet service companies. And these are Internet service companies that offer services, but they're not public cloud computing companies. The second factor has to do with cloud computing. We are now at the tipping point of cloud computing. Almost every enterprise in the world has both a cloud-first and a multi-cloud strategy. It is exactly the reason why all of the announcements that we made this year -- this quarter, this last quarter since GTC about all the new platforms that are now available in the cloud. a CSP, a hyperscaler is both -- are two things to us, therefore, a hyperscaler can be a sell to customer. They are also a cell with partner on the public cloud side of their business. Because of the richness of NVIDIA's ecosystem because we have so many Internet service customers and enterprise customers using NVIDIA's full stack. The public cloud side of their business really enjoys and values the partnership with us and the cell with relationship they have with us. And it's pretty clear now that for all of the hyperscalers, the public cloud side of their business will likely would very likely be the vast majority of their overall consumption. And so, because the world CSPs, the world's public clouds is only at the early innings of their enterprise to lifting enterprise to the cloud world it's very, very clear that the public cloud side of the business is going to be very large. And so, increasingly, our relationship with CSPs, our relationship with hyperscalers will -- will include, of course, continuing to sell to them for internal consumption but very importantly, sell with for the public cloud side.

Joseph Moore:

Great. Thank you. I wonder if you could talk to looking backward at the crypto impact. Obviously, that's gone from your numbers now, but do you see any potential for liquidation of GPUs that are in the mining network, any impact going forward? And do you foresee blockchain being an important part of your business at some point down the road?

Operator:

Your last question today comes from the line of Toshiya Hari with Goldman Sachs. Your line is now open.

Colette Kress:

Sure. When we look at our supply constraints that we have had in the past, each and every quarter, this is getting better Networking was one of our issues probably a year ago, and it has taken us probably to this quarter. and next quarter to really see our supply improved so that we can support the pipeline that we have for our customers that are -- now that's our supply. We've also made a discussion regarding our customers, supply constraints, issues when setting up a data center, even getting data center capacity has been very difficult. And therefore, that challenges them in their purchasing decisions as they're still looking for certain parts of that supply chain to come through. So that hopefully clarifies what we were talking about regarding two areas of supply. In our stock-based compensation, what we'll see, it's very difficult to predict what our stock-based compensation would be when it arrives. We have provided to our incoming employees but also once a year to our employees, and it's a single date in terms of when that is priced. So it's difficult to determine, but stock-based compensation is an important part of our employees' compensation and will continue to be. So we look at it from an overall compensation perspective. So up until now and when we do the focal, we'll see about the same size with a few additions for the reduced level of employee hiring that we have right now.

Jen-Hsun Huang:

Thanks, everyone. We are quickly adapting to the macro environment. Correcting inventory levels, offering alternative products to data center customers in China and keeping our opex flat for the next few quarters. Our new platforms are off to a great start and formed the foundation for our resumed growth. MRTX is reinventing 3D graphics with ray tracing and AI. The launch of [Inaudible] is phenomenal. Gamers waited in long lines around the world, 4090 stocks sold out quickly. Hopper, with its revolutionary transformer engine is just in time to meet the surging demand for recommender systems, large language models and generative AI. NVIDIA networking is synonymous with the highest data center throughput and enjoying record results. Oren is the world's first computing platform designed for AI-powered autonomous vehicles and robotics and putting automotive on the road to be our next multibillion-dollar platform. These computing platforms run NVIDIA AI and NVIDIA Omniverse, software libraries and engines that help the companies build and deploy AI to products and services. we this pioneering work and accelerated computing is more vital than ever. Limited by business, general purpose commuting has slowed to a crawl just as AI demands more computing. Scaling through general purchase computing alone is no longer viable, both from a cost or power standpoint. Accelerated computing is the path forward. We look forward to updating you on our progress next quarter.

Operator:

Good afternoon. My name is Regina, and I will be your conference operator today. At this time, I would like to welcome everyone to NVIDIA’s Second Quarter Earnings Call. All lines have been placed on mute to prevent any background noise. After the speakers’ remarks, there will be question-and-answer session. [Operator Instructions] [Technical Difficulty]

Colette Kress:

Thanks, Simona. This was a challenging quarter. Total revenue of $6.7 billion was down 19% sequentially and up 3% year-on-year, below the $8.1 billion outlook we provided on our last earnings call. As we indicated in our pre-announcement press release on August 8, we experienced a shortfall to our expectations driven primarily by weaker Gaming revenue. Today, we will share with you more details on our Q2 results and Q3 outlook. Starting with Gaming. Revenue of $2.04 billion was down 44% sequentially and down 33% year-on-year, reflecting challenging market conditions. As discussed in May, we expected a sequential decline in Gaming revenue due to softness in Europe related to the war in Ukraine and COVID lockdowns in China. The decline in Gaming GPU revenue was sharper than anticipated, driven by both lower units and lower ASPs. Macroeconomic headwinds across the world drove a sudden slowdown in consumer demand. We implemented programs with our Gaming channel partners to adjust pricing in the channel and to price position current high-end desktop GPUs as we prepare for a new architecture launch. As noted last quarter, we had expected cryptocurrency mining to make a diminishing contribution to gaming demand. We are unable to accurately quantify the extent to which reduced crypto mining contributed to the decline in gaming demand. While Gaming navigates significant short-term macroeconomic challenges, we believe the long-term fundamentals in Gaming remain strong. NVIDIA RTX has redefined computer graphics and is now supported by almost 300 games and applications. NVIDIA’s GeForce GPUs are the most coveted brand by gamers, representing 15 of the top 15 most popular GPUs on Steam. Gaming has emerged from the pandemic an even more popular form of entertainment and social connectivity. Estimated GeForce sell-through is up over 70% since before the pandemic, and peak concurrent users on Steam are also up more than 70% over the same time period. GeForce NOW registered members now exceed 20 million. This quarter, we added 80 more titles, including the hugely popular Genshin Impact, bringing our total to over 1,350. Moving to Professional Visualization. Revenue of $496 million was down 20% sequentially and down 4% from a year ago. A sequential increase in mobile revenue was more than offset by lower desktop revenue, particularly at the high end. As macroeconomic headwinds intensified, enterprise demand slowed and OEMs worked to reduce inventory. We expect these trends to persist in Q3. While ProViz is undergoing a near-term adjustment after doubling last year, we believe we have expanded the market opportunity over the last couple of years with AI and Omniverse workloads. We believe hybrid work is here to stay, and with it, the need for collaborative 3D design enabled by professional graphic workstations, both at home and in the office as well as in the cloud. In June, we announced a partnership with Siemens to enable the industrial levers and AI-powered digital twins, connecting Siemens Xcelerator platform to NVIDIA Omniverse. This connection opens Siemens to the vast ecosystem of NVIDIA Omniverse and NVIDIA to Siemens ecosystem of the world’s largest industries. Earlier this month at SIGGRAPH, the premier computer graphics conference, we announced advancements to several foundational technologies of the metaverse, defined as the 3D version of the Internet. First, NVIDIA Omniverse Avatar Cloud Engine will enable businesses to create and deploy assistants and avatars, transforming interactions across a range of industries. We also unveiled 11 new Omniverse connectors, bringing the total number of connectors to the Omniverse USD ecosystem to 112. And finally, we released SDKs for the new field of neural graphics, which intertwine AI and graphics to help automate the creation of virtual world. Moving to Automotive. Revenue of $220 million increased 59% sequentially and 45% from the year-ago quarter. Strong growth was driven by auto AI solutions, which include AI cockpit and self-driving revenue, with particular strength in self-driving as new energy vehicle design wins ramp into volume. We believe Q2 was an inflection point for our automotive revenue as NVIDIA Orin has great momentum. During the quarter, we announced rollout plans of new vehicles from OEM partners, NIO, Li Auto, JIDU and Human Horizons as well as Pony.ai’s line of self-driving trucks and robotaxis, all built on NVIDIA DRIVE. Looking forward, we expect our $11 billion automotive design win pipeline to translate to continued growth. Moving to Data Center. Revenue of $3.81 billion grew 1% sequentially and 61% year-on-year. Although a record, this was somewhat short of our expectations as we were impacted by supply chain disruptions. Revenue from hyperscale customers nearly doubled year-on-year. Sequentially, sales to North America hyperscale and cloud computing customers increased but were more than offset by lower sales to China hyperscale customers affected by domestic economic conditions. Vertical industries grew both, sequentially and year-on-year. Key workloads driving growth include natural language processing, recommender systems, autonomous vehicle fleet, data processing and training and cloud graphics. Let me share a couple of customer examples. Pinterest transitioned to 100x larger recommender models by moving its inference from CPUs to NVIDIA GPUs. Its ability to deploy a higher-quality model at high throughput and low latency resulted in a 16% increase in engagement, a critical metric for the company, which has over 400 million users and 300 billion images. And Tesla recently upgraded its supercomputer to use over 7,000 A100 GPUs for autopilot training. From a product perspective, networking led growth this quarter with strong demand from our high-speed Ethernet adapters and design win momentum toward next-generation adopters, including the ConnectX-6 and ConnectX-7. We also see growing interest from cloud service providers for our new Spectrum-4 400 gigabit per second Ethernet networking platform. Additionally, we are ramping into the upcoming launches of our next generation platforms. The Hopper architecture flagship H100 data center GPU is in production. Grace is our first CPU. Top computer makers, including Dell, HPE, Inspur, Lenovo and Supermicro are adopting the new NVIDIA Grace CPU Superchip and Grace Hopper Superchip to build the next generation of supers. 72% of the systems on the latest top 500 list of the world’s fastest supercomputers are powered by NVIDIA, including 31 of 39 new systems. NVIDIA’s [Indiscernible] Selene supercomputer ranks at number 8 in the top 500 and is the world’s fastest enterprise supercomputer. Moreover, 22 of the top 30 systems on the Green 500 list of the most energy-efficient supercomputers are powered by NVIDIA. Significant advances in software technologies are key to our platform performance. In the past two years, our A100-based platform has delivered 6x more performance as measured by the MLPerf industry benchmark, largely through new software technologies and optimizations. Last month, we announced an update to the NeMo Megatron framework that can speed up the training of large language models by up to 30%, improving a multi-hundred million dollar AI infrastructure by 30% translates to significant value for customers. LLM are one of the most important neural networks today, ranging in size from tens of billions to over 1 trillion parameters. Learning from text, they can be used for real-time content generation, tech summarization, customer service chat box and question answering for conversational AI interfaces. Currently, these capabilities are available to early access customers to run on NVIDIA DGX SuperPOD and NVIDIA DGX Foundry as well as in Microsoft Azure cloud with other platforms available soon. We are working with the industry leaders in large language models, a very active and exciting space of AI. Moving to the rest of the P&L. GAAP gross margin was 43.5%, and non-GAAP gross margin was 45.9%. Gross margin includes $1.22 billion in charges for inventory and related reserves based on revised expectations of future demand and $122 million for warranty reserves. These charges incurred in the quarter reflect purchase commitments that we made during the time of severe component shortages and our current expectation of ongoing macro uncertainty. We believe our long-term gross margin profile is intact. GAAP operating expenses were up 36% from a year ago and down 32% sequentially as Q1 included a $1.35 billion acquisition termination charge related to the Arm transaction. Non-GAAP operating expenses were up 38% from a year ago and up 9% sequentially. These increases were driven primarily by employee growth costs, as well as increases in salaries to support our employees during this high inflationary environment, and engineering development of new products coming to market. We have slowed operating expense growth, balancing investments for long-term revenue growth while managing near-term profitability. Our full year non-GAAP OpEx is expected to grow over 30%. During the first half of fiscal 2023, we returned $5.5 billion to shareholders in the form of share repurchases and cash dividends. We plan to continue share repurchases. We have nearly $12 billion remaining under our authorization through December of 2023. Let me turn to the outlook for the third quarter of fiscal ‘23. We expect Gaming and ProViz revenue to decline sequentially as OEMs and channel partners reduce inventory levels to align with current levels of demand and prepare for our new product generation. We expect that decline to be partially offset by sequential growth in data center and automotive. Revenue is expected to be $5.9 billion plus or minus 2%. GAAP and non-GAAP gross margins are expected to be 62.4% and 65%, respectively, plus or minus 50 basis points. GAAP operating expenses are expected to be approximately $2.59 billion. Non-GAAP operating expenses are expected to be approximately $1.82 billion. GAAP and non-GAAP other income and expenses are expected to be an expense of approximately $10 million, excluding gains and losses on nonaffiliated investments. GAAP and non-GAAP tax rates are expected to be 9.5%, plus or minus 1%, excluding any discrete items. Capital expenditures are expected to be approximately $550 million to $600 million. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight upcoming events for the financial community. We will be attending the Jefferies Conference in Chicago on August 30th and the Goldman Sachs Conference in San Francisco on September 12th. And we will be holding a financial analyst Q&A with management following Jensen’s GTC keynote on September 20th. Our earnings call to discuss the results of our third quarter of fiscal 2023 is scheduled for Wednesday, November 16th. We will now open the call for questions. Operator, can you assist? Would you please poll for questions?

C.J. Muse:

I think the question we all have is what is normalized revenues for gaming for you guys? Obviously, this is a challenge to you as well. But curious how you’re thinking about it today. Is the fiscal ‘20 recovery post the first half ‘19 correction an appropriate framework, or was that inflated by crypto as well? And I guess, as part of that, how do we think about the cascading in of the new product cycle? And is there potential for future reserves needed to be taken if gaming does not meet your new updated outlook? Thanks so much.

Jensen Huang:

Yes. Thanks, Colette. C.J., our sell-through is off the highs in the beginning of the year, but it’s still very solid. In fact, sell-through is -- has increased 70% since pre-COVID, pre-pandemic. And so, it’s very clear that gaming is -- the fundamentals of gaming are strong, and this medium is really doing well. Not to mention the gaming platforms are being used -- our gaming PCs are being used for influencers, people sharing content, creating content, V bloggers, VTuber, there’s all kinds of new ways of engaging and spending time with video games. Our strategy is to reduce the sell-in -- reduce the sell-in this quarter, next quarter to let channel inventory correct. Obviously, we’re off the highs, and the macro condition turned sharply worse. And so, our first strategy is to reduce sell-in in the next couple of quarters to correct channel inventory. We’ve also instituted programs to price position our current products to prepare for next-generation products. Ampere is the most popular GPU we’ve ever created. It is in the top 15 most popular gaming GPUs on Steam. And it remains the best GPUs in the world, and it will be very successful for some time. However, we do have exciting new next-generation coming and it’s going to be layered on top of that. And so, we’ve taken -- we’ve done two things. We’ve reduced sell-in to let channel inventory correct and we’ve implemented programs with our partners to price position the products in the channel in preparation for our next generation. All of this we anticipate were working towards a path to being in a good shape going into next year. Okay? So, that’s what our game plan is.

Vivek Arya:

Actually, I just wanted to clarify, Jensen. So, should we assume that a gaming sell-in will kind of stay at these levels into your Q1 or Q2, or depending on new product launch, it might recover? So just wanted to make sure our baseline assumption is set there. And then, my question is actually similar on the data center. Sales are pretty strong right now, but there is a concern that data center CapEx could be the next shoe to drop in this rolling correction in semiconductors. I’m curious, what’s your sense of utilization of your data center shipments? And what is the risk that there could be a correction in the data center given some of the macro caution expressed by some of the hyperscaler and enterprise customers?

Operator:

Your next question will come from the line of Matt Ramsay with Cowen.

Colette Kress:

Great. Thanks for the question. And it’s kind of a little bit of an add-on to some of the statements that Jensen was discussing regarding our supply chain and what we’re seeing today. Our supply chain during the quarter really was quite difficult, was quite challenging to work through. Our platforms, including HDX, networking chips, cables, switches, were very important to the customers. It’s not just about us selling the GPUs. So even though customers orders components themselves, they’re looking from us what we may refer to as kits, kits that go with those GPUs for them to stand up their data centers. We also experienced supply disruptions internally with our logistics and our component availability. Some of our supply arrived very late in the quarter. We had very little time from a logistics and availability to get those things out. Customers were impacted as well by availability of key third-party other components that we weren’t offering, which were slowing down some of their deployments. So what we did in our Q2 orders that couldn’t be delivered in Q3, given that some of these supply constraints existed, and we had Q3 demand where we did have supply in Q2. So we worked with customers to optimize that supply and demand, and that’s what we’ve disclosed to you.

Operator:

Your next question will come from the line of Toshiya Hari with Goldman Sachs.

Colette Kress:

Thanks for the question. So our inventory charges, as we commented, we’re taking a thorough look of not only this last quarter as well as the quarter that we’re guiding, but looking over the long term of what we need for demand and then what we had in terms of supply. Remember, we had purchased this very early on in the year as we needed to, to drive the commitment of the supply that we already have. And so what is happening here for the data center, we had great, high expectations. We still have very strong, solid growth projection for data center as well. We’re going to take this opportunity for some of the prior architecture pieces to write down those given what we see as just a change in terms of our expectations going forward. So you are correct. There are also pieces in there for gaming. We have written down some silicon and chips as the macroeconomic conditions and you get ready for our future product launches take into account, but there’s also components, services and capacity in some of the other drivers that are incorporated in those write-downs.

Operator:

Your next question will come from the line of Aaron Rakers with Wells Fargo.

Colette Kress:

Yes. Thanks for the question. And that’s a pretty good understanding of our guidance. And we do expect, yes, gaming to decrease, not in the dollar amount that it decreased between Q1 and Q2. So that may be of our 2 areas of the decline, our gaming and ProViz, that may be about 3/4 of it associated with that gaming. And then Professional Visualization would probably be about 1/4 of the 2 areas that will decline. Our Data Center yes, we do expect it to grow. It may grow about what we just saw between Q1 and Q2. We’ll continue to look at it. There may be some more opportunity there. And automotive, very similar to our thoughts at the very beginning of the quarter, we are expecting continued growth through each of the quarters of this fiscal year. We felt that Q2 was an inflection point. So we’ll continue to grow into Q3 and hopefully Q4 going forward. I’ll turn it over to Jensen to see his thoughts in terms of Hopper, what Hopper brings to us in Q4 and expectations.

Operator:

Your next question will come from the line of Atif Malik with Citi.

Colette Kress:

So within our Q2 results, we have been continuing to improve our supply for networking. We have a lot of important products that the CSP needs, many of our customers’ needs, and we have been working to really improve that supply. And we were able to set very strong growth in terms of networking, both sequentially and year-over-year. And as we move into the next quarter, we’re going to have to see which is going to be growth larger. We’re just going to have to take a lot once we finish that quarter. But our supply for compute is here. But as we’ve discussed, sometimes it’s important that they have many of our other components that we provide in networking at the same time that we are providing the GPUs. So sometimes those are very important for us to deliver together. So we always have to keep that in mind. So it’s not always supply constrained, but there are certain parts of it that are.

Joseph Moore:

Great. I wanted to ask why did the supply constraints hit you guys so hard this quarter? I mean you’ve done such a great job in the last couple of years outgrowing really everybody in a very challenging supply chain environment. It hasn’t tripped you up at all. And it seems like now it’s kind of hitting you fairly hard at a time when, in other cases, it’s kind of easing. So I’m just kind of curious, what is it about the timing and how long does it sort of take? Or is that a Hopper issue? Is it related to other components? As you talked about, how long does it take to clear those issues up?

Operator:

Your next question will come from the line of Stacy Rasgon with Bernstein.

Colette Kress:

Yes, Stacy. Regarding our inventory charges that we had, when you think about what we have in supply, whether it be chips, components, whether it be memory. Remember, a lot of these things can be used interchangeably across the 2. Additionally, the value of our inventory for data center is much different than the value of what we have for gaming from an overall cost perspective. So we’re creating capacity opportunity, putting together all of those systems in terms of data center. It is prior architectures. Absolutely, this is not a question regarding anything of our future products coming to market. Nothing on the inventory provision has to do with that. So we took this as looking at the macroeconomic conditions, as we’ve discussed. Our expectations, our plans were higher. They’re still quite solid that we see in demand both for gaming as well as solid for data center, and that will continue. But we did have to just take a rightsizing of that note.

Operator:

Your next question will come from the line of Srini Pajjuri with SMBC Nikko.

Colette Kress:

Yes. Thanks for the question. So our gross margins outside of the inventory charges in Q2 as well as going into Q3 is really about our sales mix that we have and probably also to understand that our sales mix in the next quarter for GPUs is not in the high end. And so that has impacted our gross margin as we move into Q3. You are correct. We do expect that data center will assist in our gross margins but we also have growth plans in auto. Auto is below our company average, and so that will tend to offset some of those upper bound things that we will see in terms of data center. From time to time, there’s always a small amount of scraps that we will have in our gross margin estimates. So nothing material is planned. But there is small scraps that may occur from quarter-to-quarter that are included in our gross margins.

Tim Arcuri:

Colette, I had a clarification and then a question. So my clarification is whether October gross margins are benefiting at all from the sale of previously written down inventory. And then my question is whether you can give us the enterprise cloud split in data center because it sounds like the mix shifted more towards enterprise in July. And I think investors might want to see that as risky in the face of enterprise clearly slowing. So I’m wondering if you can give us that.

Operator:

Your next question will come from the line of Harlan Sur with J.P. Morgan.

Colette Kress:

So our Pro Visualization business, just in such a short time, they were in short supply and really trying to feed so much of the industry’s desire to both refresh and attracted to the new RTX workstations that were there. We were fueling both mobile as well as desktop for that market and growing quite nicely. Remember, all of last year, doubled the size of ProViz in 1 year. So right now, the OEMs are concentrating on their levels of inventory. We want to make sure that they can get through that inventory. But it -- keep in mind, very similar to our discussion on gaming, there is still solid demand. There is still solid demand. We just have to correct some of the inventory, but we still see both the opportunities that we’ve created for the market for these to be long-standing. And your second question, remind me again?

Operator:

Your final question will come from the line of Ross Seymore with Deutsche Bank.

Jensen Huang:

I would say that without crypto dynamic, the mix would go down. However, the overall trend long term, the ASP is drifting up. And the way to think about that is a game console, when -- my first game console was $99. Lately, game consoles are selling for about $599. And the reason for that is because it’s more useful than ever. You use your gaming console for your greatest form of entertainment, and you use it for a very, very long time. And GeForce essentially is a game console inside your PC. And we’ve always believed that the ASP of GeForce should drift towards the average selling price of a game console. And so it should be something along the lines of $500 or so roughly at this time. We also have GeForce in the cloud. And because GeForce in the cloud is hosting many gamers simultaneously, it tends to want to be a much more powerful GeForce. And so our cloud gaming GeForce tends to be -- our cloud gaming graphics tends to be a much higher end. And so -- and then, of course, there’s the design aspect of it. Most designers and most creators are able to use GeForce these days. And they use their PC to create content, and much of that content goes into video games and/or they’re using video games to create their artistic content. And so the GeForce is not just for gaming for them. The GeForce is essentially their creative work station as well. And so there are several dynamics that are causing the ASP of GeForce to go up, and we’ve been seeing this trend for several years now.

Jensen Huang:

Thanks, everyone. We’re navigating our supply chain transitions in a challenging macro environment. In Gaming, our partners and ecosystem are responding to a sudden slowdown in consumer demand and correcting channel inventory. Still, the fundamentals of gaming are strong. We’ll get through this over the next few months and go into next year with our new architecture. I look forward to telling you more about it at GTC next month. In Data Center, AI where computers are helping us write software that was impossible before is driving a computing revolution and transforming every industry. NVIDIA’s leadership in full stack data center scale, accelerated computing has made us the ideal partner for companies racing to leverage the power of AI. Even with the current macroeconomic headwinds, demand for our data center products have never been stronger. The next wave of computing is coming. With AI and 3D graphics advances, developers will extend the Internet with virtual world overlays that connect to the physical world. This next evolution of the Internet is called metaverse. We created Omniverse to connect the digital and physical world and be an open platform for creating and operating metaverse applications. The immediate applications for Omniverse span product design, manufacturing and operations. Omniverse is off to a great start. Our automotive revenue is inflecting, and we expect it to be our next $1 billion business. Autonomous driving is one of the biggest challenges AI can solve, and computing opportunity for us spans the data center to the car. Autonomous driving will transform the auto industry into a tech industry. Automotive is one of the first to transform into a software-defined tech industry that all industries will be. We’re building NVIDIA AI and NVIDIA Omniverse to be the engines for the world’s enterprise to become software-defined, AI-powered technology companies. I look forward to next month’s GTC conference, where we will share new advances of RTX reinventing 3D graphics and gaming. AI’s continuing breakthroughs and building the metaverse, the next evolution of the internet. So join us. We look forward to updating you on our progress next quarter. Thank you.

Operator:

Good afternoon. My name is David, and I'll be your conference operator today. At this time, I'd like to welcome everyone to NVIDIA's first quarter earnings call. Today's conference is being recorded. All lines have been placed on mute to prevent any background noise. After the speaker’s remarks, there will be question-and-answer session. [Operator Instructions] Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. We delivered a strong quarter driven by record revenue in both Data Center and Gaming with strong fundamentals and execution against a challenging macro backdrop. Total revenue of $8.3 billion was a record, up 8% sequentially and up 46% year-on-year. Data Center has become our largest market platform, and we see continued strong momentum going forward. Starting with Gaming. Revenue of $3.6 billion rose 6% sequentially and 31% year-on-year, powered by the GeForce RTX 30 Series product cycle. Since launching in the fall of 2020, the RTX 30 Series has been our best Gaming product cycle ever. The gaming industry has grown tremendously with 100 million new PC gamers added in the past two years according to Newzoo. And NVIDIA RTX has set new standard for the industry with demand from both first-time GPU buyers as well as those upgrading their PCs to experience the 250-plus RTX-optimized games and apps, double from last year. We estimate that almost a-third of the GeForce Gaming GPU installed base is now on RTX. RTX has brought tremendous energy into the gaming world, and has helped drive a sustained expansion in our higher-end platforms and installed base with significant runway still ahead. Overall, end demand remained solid though mixed by region, and demand in Americas remained strong. However, we started seeing softness in parts of Europe related to the war in the Ukraine and parts of China due to the COVID lockdowns. As we expect some ongoing impact as we prepare for a new architectural transition later in the year, we are projecting Gaming revenue to decline sequentially in Q2. Channel inventory has nearly normalized and we expect it to remain around these levels in Q2. The extent in which cryptocurrency mining contributed to Gaming demand is difficult for us to quantify with any reasonable degree of precision. The reduced pace of increase in Ethereum network hash rate likely reflects lower mining activity on GPUs. We expect a diminishing contribution going forward. Laptop Gaming revenue posted strong sequential and year-on-year growth, driven by the ramp of the NVIDIA RTX 30 Series lineup. With this year's spring refresh and ahead of the upcoming back-to-school season, there are now over 180 laptop models featuring RTX 30 Series GPUs and our energy-efficient thin and light Max-Q technologies, up from 140 at this time last year. Driving this growth are not just gamers, but also the fast-growing category of content creators for whom we offer dedicated NVIDIA studio drivers. We've also developed applications and tools to empower artists from Omniverse for advanced 3D and collaboration to broadcast for live streaming to Canvas for painting landscapes with AI. The creator economy is estimated at $100 billion and powered by 80 million individual creators and broadcasters. We continued to build out our GeForce NOW cloud gaming service. Gamers can now access RTX 3080 class streaming, our new top-tier offering with subscription plans of $19.99 a month. We added over 100 games to the GeForce NOW library, bringing the total to over 1,300 games. And last week, we launched Fortnite on GeForce NOW with touch controls for mobile devices, streaming through the Safari web browser on iOS and the GeForce NOW Android app. Moving to Pro Visualization. Q1 revenue was $622 million was down sequentially 3% and up 67% from a year ago. Demand remains strong as enterprises continued to build out their employee's remote office infrastructure to support hybrid work. Sequential growth in the mobile workstation GPUs was offset by lower desktop revenue. Strong year-on-year growth was supported by the NVIDIA RTX Ampere architecture product cycle. Top use cases include digital content creation at customers such as Sony Pictures Animation and medical imaging at customers such as Medtronic. In just its second quarter of general availability, our Omniverse enterprise software is being adopted by some of the world's largest companies. Amazon is using Omniverse to create digital twins to better optimize warehouse design and flow and to train more intelligent robots. Kroger is using Omniverse to optimize store efficiency with digital twin store simulation. And PepsiCo is using Omniverse digital twins to improve the efficiency and environmental sustainability of the supply chain. Omniverse is also expanding our GPU sales pipeline, driving higher end and multiple GPU configurations. The Omniverse ecosystem continues to rapidly expand with third-party developers in the robotics, industrial automation, 3D design and rendering ecosystems developing connections to Omniverse. Moving to automotive. Q1 revenue of $138 million, increased 10% sequentially and declined 10% from the year ago quarter. Our DRIVE Orin SoC is now in production and kicks off a major product cycle with auto customers ramping in Q2 and beyond. Orin has great traction in the marketplace with over 35 customer wins from automakers, truck makers and robotaxi companies. In Q1, BYD, China's largest EV maker and Lucid an award winning EV pioneer were the latest to announce that they are building their next-generation fleets on DRIVE Orin. Our automotive design win pipeline now exceeds $11 billion over the next six years, up from $8 billion just a year ago. Moving to Data Center. Record revenue of $3.8 billion grew 15% sequentially and accelerated to 83% growth year-on-year. Revenue from hyperscale and cloud computing customers more than doubled year-on-year, driven by strong demand for both external and internal workloads. Customers remain supply constrained in their infrastructure needs and continue to add capacity as they try to keep pace with demand. Revenue from vertical industries grew a strong double-digit percentage from last year. Top verticals driving growth this quarter include consumer Internet companies, financial services and telecom. Overall, Data Center growth was driven primarily by strong adoption of our A100 GPU for both training and inference with large volume deployments by hyperscale customers and broadening adoption across the vertical industries. Top workloads includes recommender systems, conversational AI, large language models and cloud graphics. Networking revenue accelerated on strong broad-based demand for our next-generation 25, 50 and 100-gig ethernet adapters. Customers are choosing NVIDIA's networking products for their leading performance and robust software functionality. In addition, networking revenue is benefiting from growing demand for DGX super pods and cross-selling opportunities. Customers are increasingly combining our compute and networking products to build what are essentially modern AI factories with data as the raw material input and intelligence as the output. Our networking products are still supply constrained though we expect continued improvement throughout the rest of the year. One of the biggest workloads driving adoption of NVIDIA AI is natural language processing, which has been revolutionized by transformer based models. Recent industry breakthroughs traced to transformers include; large language models like GPT-3, NVIDIA Megatron BERT for drug discovery and DeepMind AlphaFold for a protein structure prediction. Transformers allow self-supervised learning without the need for human labeled data. They enable unprecedented levels of accuracy for TAF such as text generation, translation, summarization and answering questions. To do that, Transformers use enormous training data sets and very large networks well into the hundreds of billions of parameters. To run these giant models without sacrificing low inference times, customers like Microsoft are increasingly deploying NVIDIA AI, including our NVIDIA Ampere architecture-based GPUs and full software stack. In addition, we are seeing a rising wave of customer innovation using large language models that is driven by increased demand for NVIDIA AI and GPU instances in the cloud. At GTC, we announced our next-generation Data Center GPU, the H100 based on the new or upper architecture. Packed with 80 billion transistors, H100 is the world's largest, most powerful accelerator, offering an order of magnitude leap in performance over the A100. We believe H100 is hitting the market at the perfect time. H100 is ideal for advancing large language models and deep recommender systems the two largest scale AI workloads today We are working with leading server makers and hyperscale customers to qualify and ramp H100. As well as the new DGX H100 AI supercomputing system will ramp in volume late in the calendar year. Building on the H100 product side, we are on track to launch our first ever Data Center CPU, Grace, in the first half of 2023. Grace is the ideal CPU for AI factories. This week at COMPUTEX, we announced that dozens of server models based on Grace will be brought to market by the first wave of system builders, including ASUS, Foxconn, Gigabyte, QCT, Supermicro and Wiwynn. These servers will be powered by the NVIDIA Grace CPU Super Chip, which features two CPUs and the Grace Upper Super Chip, which pairs an NVIDIA Hopper GPU with an NVIDIA Grace CPU in an integrated model. We've introduced new reference designs based on Grace for the massive new workloads of next-generation data centers. CGX for cloud graphics and gaming, OVX for digital twins or Omniverse and HGX for HPC and AI. These server designs are all optimized for NVIDIA's rich accelerated computing software stacks and can be qualified as part of our NVIDIA certified systems lineup. The enabler for the Grace Hopper and Grace Super Chips is our ultra energy-efficient, low-latency, high-speed memory coherent interconnect NVLink, which scales from die to die, chip to chip and system to system. With NVLink, we can configure Grace and Hopper to address a broad range of workloads. Future NVIDIA chips, the CPUs, GPUs, DPUs, NICs and SoCs will integrate NVLink just like Grace Hopper based on our world-class SERDES technology. We're making NVLink open to customers and partners to implement custom chips that connect to NVIDIA's platforms. In networking, we're kicking off a major product cycle with the introduction of Spectrum-4, the world's first 400-gigabit per second end-to-end Ethernet networking platform, including the Spectrum-4 Switch, ConnectX-7 SmartNIC, Bluefield-3 DPU and the DOCA software. Built for AI and video Spectrum 4 arrives as data centers are growing exponentially and demanding extreme performance, advanced security and powerful features to enable high-performance advanced virtualization and simulation at scale. Across our businesses, we are launching multiple new GPU, CPU, DPU and SOC quarters over the coming quarters, with a ramp in supply to support the customer demand. Moving to the rest of the P&L, GAAP gross margin for the first quarter was 65.5% and non-GAAP gross margin was up 67.1%, up 90 basis points from a year ago, and up 10 basis points sequentially. We have been able to offset rising costs and supply chain pressures. We expect to maintain gross margins at current levels in Q2. Going forward, as new products ramp and software becomes a larger percent of revenue, we have opportunities to increase gross margins longer term. GAAP operating margin was 22.5%, impacted by a $1.35 billion acquisition termination charge related to the ARM transaction. Non-GAAP operating margin was 47.7%. We are closely managing our operating expenses to balance the current macro environment with our growth opportunities, and we've been very successful in hiring so far this year and are now slowing to integrate these new employees. This also enables us to focus our budget on taking care of our existing employees as inflation persist. We are still on track to grow our non-GAAP operating expenses in the high 20s range this year. we expect sequential increases to level off after Q2 as the first half of the year includes a significant amount of expenses related to the bring-up of multiple new products, which should not reoccur in the second half. During Q1, we repurchased $2 billion of our stock. Our Board of Directors increased and extended our share repurchase program to repurchase an additional common stock up to a total of $15 billion through December 2023. Let me now turn to the outlook for the second quarter of fiscal 2023. Our outlook assumes an estimated impact of approximately $500 million relating to Russia and China COVID lockdowns. We estimate the impact of lower sell-through in Russia and China to affect our Q2 Gaming sell-in by $400 million. Furthermore, we estimate the absence of sales to Russia to have a $100 million impact on Q2 in Data Center. We expect strong sequential growth in Data Center and Automotive to be more than an offset by the sequential decline in Gaming. Revenue is expected to be $8.1 billion, plus or minus 2%. GAAP and non-GAAP gross margins are expected to be 65.1% and 67.1%, respectively, plus or minus 50 basis points. GAAP operating expenses are expected to be $2.46 billion. Non-GAAP operating expenses are expected to be $1.75 billion. GAAP and non-GAAP other income and expenses are expected to be an expense of approximately $40 million, excluding gains and losses on non-affiliated investments. GAAP and non-GAAP tax rates are expected to be 12.5% plus or minus 1%, excluding discrete items. And capital expenditures are expected to be approximately $400 million to $450 million. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight the upcoming events for the financial community. We will be attending the BofA Securities Technology Conference in person on June 7, where Jensen will participate in a keynote fireside chat. Our earnings call to discuss the results of our second quarter of fiscal 2023 is scheduled for Wednesday, August 24. We will now open the call for questions. Operator, can you please poll for questions? Thank you.

C.J. Muse:

Yes, good afternoon. Thank you for taking the question. I guess would love to get an update on how you're thinking about the Gaming cycle from here. The business has essentially doubled over the last two years. And now we've got some crosswinds with crypto falling off, channel potentially clearing ahead of a new product cycle. You talked about macro challenges. But at the same time, only a third of the installed base has RTX and we're moving out from under supply. So we'd love to hear your thoughts from here once we get beyond kind of the challenges around COVID lockdown in the July quarter? How are you thinking about Gaming trends?

Operator:

Next, we'll go to Matt Ramsay with Cowen. Your line is open.

Jensen Huang:

Yes. I appreciate the question. The -- let's see, the first point that you made is a very big point. The next generation of servers that are being teed up right now are all Gen 5. The I/O performance is substantially higher than what was available before. And so, you're going to see a pretty large refresh as a result of that. Brand-new networking cards from our company and others. Gen 5, of course, drives new platform refresh. And so, we're perfectly timed to ramp into the Gen 5 generation with Hopper. There are a lot of different system configurations you want to make. If you take a step back and look at the type of systems that are necessary for data processing, scientific computing, machine learning and training, inference done in the cloud for hyperscale nature, done on-prem for enterprise computing, done at the edge. Each one of these workloads and deployment locations, the way that you manage would dictate a different system architecture. So there isn't one size that fits all, which is one of the reasons why it's so terrific that we support PCI Express, that we innovated chip-to-chip interconnect for diverse -- before anybody else did, this is now some seven years ago, we're in our fourth generation of NVLink that allows us to connect two chips next to each other, two dies, two chips, two modules, two SXM modules to two systems to multiple systems. And so our coherent chip-to-chip link, NVLink has made it possible for us to mix and match chips, dies, packages, systems and all of these different types of configurations. And I think that, over time, you're going to see even more types of configurations. And the reason for that has to do with a couple of very important new type of data centers that are emerging. And you're starting to see that now with fairly large installations, infrastructures with NVIDIA, HPC and NVIDIA AI. These are really AI factories where you're processing the data, refining the data and turning that data into intelligence. These AI factories are essentially running one major workload and they're running at 24/7. Deep recommender systems is a good example of that. In the future, you're going to see large language models essentially becoming a platform themselves. That would be running 24/7, hosting a whole bunch of applications. And then on the other end, you're seeing data centers at the edge that are going to be robotics or autonomous data centers that are running 24/7. They are going to be running in factories and retail stores and warehouses, logistics warehouses, all over the world. So these two new types of data centers are just emerging, and they also have different architectures. So I think the net of it all is that our ability to support every single workload because we have a universal accelerator, running every single workload from data processing to data analytics to high-performance computing to training to inference that we can support Arm and x86 that we support PCI Express to Multisystem NVlink to multi-chip NVLink to multi-die NVLink, that capability for us is -- makes it possible for us to really be able to serve all of these different segments. With respect to vertical integration, I think that system integration, the better way of maybe saying that is that system integration is going to come in all kinds of different ways. We're going to do semi-custom chips as we've done with many companies in the past, including Nintendo. We'll do semi-custom chiplets as we do with NVLink. NVLink is open to our partners. And they could bring it to any fab and connect it coherently into our chip. We could do multi module packages. We could do multi-package systems. So there's a lot of different ways to do system integration.

Stacy Rasgon:

Hi, guys. Thanks for taking my questions. I wanted to follow up on the sequential. So Colette, I know you said the $500 million was a $400 million hit to Gaming and a $100 million hit to data. So I'm assuming that -- that doesn't mean the Gaming is down $400 million. I mean it's Gaming -- do you see Gaming actually down more than the actual Russia and lockdown hit. And I guess just how do I think about the relative sequentials of the businesses in light of those constraints that you guys are facing?

Operator:

Next, we'll go to Mark Lipacis with Jefferies. Your line is open.

Jensen Huang:

Yeah. Thanks, Mark. We had a record Data Center business this last quarter. We expect to have a record, another record quarter this quarter, and we're fairly enthusiastic about the second half. AI and data-driven machine learning techniques for writing software and extracting insight from the vast amount of data that companies have is incredibly strategic to all the companies that we know. Because in the final analysis, AI is about automation of intelligence and most companies are about domain-specific intelligence. We want to produce intelligence. And there are several techniques now that have been created to make it possible for most companies to apply their data to extract insight and to automate a lot of the predictive things that they have to do and do it quickly. And so I think the trend that you hear other people are experiencing about machine learning, data analytics, data driven insights, artificial intelligence. However, it's described, it's all exactly the same thing. And it's sweeping just about every industry and every company. Our networking business is also highly supply constrained. Our demand is really, really high. And it requires a lot of components aside from just our chips. Components and transceivers and connectors and cables. And just -- it's a really -- it's a complicated system, the network, and there are many physical components. And so the supply chain has been problematic. We're doing our best and our supply has been increasing from Q4 to Q1. We're expecting it to increase in Q2 and increase in Q3 and Q4. And so we're really, really grateful for the support from the component industry around us, and we'll be able to increase that. With respect to software, there are two, well, first of all, there are all kinds of machine learning models, computer vision, speech AI, natural language understanding, all kinds of robotics applications, the most -- probably the largest, the most visible one is self-driving cars, which is essentially a robotic AI. And then recently, this incredible breakthrough from an AI model called Transformers that has led to really, really significant advances in natural language understanding. And so they're all these different types of models. There are thousands and thousands of species of AI models and used in all these different industries. One of my favorite, I'll just say it very quickly and I'll answer that question about the software. One of my favorites is using Transformers to understand the language of chemistry or using transformers and using AI models to understand the language of proteins, amino acids, which is genomics. To apply AI to understand -- to recognize the patterns, to understand the sequence and essentially understand the language of chemistry and biology is a really, really important breakthrough. And all of this excitement around synthetic biology, much of it stands back to the – some of these inventions. But anyhow, all of these different models need an engine to run on. And that engine is called NVIDIA AI. In the case of hyperscalers, they can cobble together a lot of open source and we provide a lot of our source to them and a lot of our engines to them for them to operate their AI. But for enterprises, they need someone to package it together and be able to support it and refresh it, update it for new architecture, support old architectures in their installed base, etcetera, and all the different use cases that they have. And so that engine is called NVIDIA AI. It's almost like a sequel engine, if you will. And except this is an engine for artificial intelligence. There's another engine that we provide and that engine is called Omniverse and it's designed for the next wave of AI, where artificial intelligence has to not just manipulate information like recommender systems and conversational systems and such. But it has to interact with physical systems. Whether it's interacting with physics directly, meaning robotics or being able to automate physical systems like heat recovery steam generators, which is really important today. And so Omniverse is designed to be able to sit at that interface, that intersection between simulation and artificial intelligence, and that's what Omniverse is about. Omniverse has now – let’s see some -- we're still early in the deployment of Omniverse for commercial license. It's been a couple of quarters now since we've released Omniverse enterprise. And I think, at this point, we have 10% of the world's top 100 companies that are already customers, licensing customers, substantially more who we're evaluating. I think it's been downloaded nearly 200,000 times. It has been tried in some 700 companies. And Colette highlighted some of the companies, you might see some of the companies that are using it in all kinds of interesting applications at GTC. And so, I fully expect that the NVIDIA AI engine, the Omniverse engine, are going to be very successful for us in the future and contribute greatly to our earnings.

Vivek Arya:

Thanks. Just wanted to clarify, Colette, if your Q2 outlook includes any destocking benefits from the new products that you're planning to launch this year? And then my question is gentleman for you. You're still guiding Data Center to a very strong, I think, close to 70% or so year-on-year growth, despite all the headwinds. Are you worried at all about all the headlines about the slowdown in the macro economy? I like is there any cyclical impact on Data Center growth that we should keep in mind as we think about the second half of the year?

Jensen Huang:

Thanks. Our Data Center demand is strong and remains strong. Hyperscale and cloud computing revenues, as you mentioned, has grown significantly. It's doubled year-over-year. and we're seeing really strong adoption of A100. A100 is really quite special and unique in the world of accelerators. And this is one of the really, really great innovations as we extended our GPU from graphics to CUDA to Tensor Core GPUs. It's now a universal accelerator. And so you could use it for data processing for ETL, for example, extract, transform and load. You could use it for database acceleration. Many sequel functions are accelerated on NVIDIA GPUs. We accelerate Rapids, we accelerate which is the Python version a Data Center scale version of Pandas, we accelerate Spark 3.0. And so from database queries to data processing, to extraction, and transform and loading of data before you do training and inference and whatever image processing or other algorithmic processing you need to do can be fully accelerated on A100. And so we're seeing great success there. on at the core and closer to what is happening today, you're seeing several different very important new AI models that are being invested in at very, very large scale and with great urgency. You probably have heard about Deep Recommender Systems. This is the economic engine, the information filtering engine of the Internet, if not for the recommender system, it would be practically impossible for us to enjoy our Internet experience shopping experience with trillions of things that are changing in the world every day constantly and be able to use your three-inch phone to even engage the Internet. And so all of that magic is made possible by this incredible thing call a recommender system second thing is conversational AI. You're seeing chat bots and website customer service, even live customer service being now supported by AI, conversational AI has an opportunity to enhance the customer service on the one hand. On the other hand, supplement for a lot of labor shortage. And then the third is this groundbreaking piece of work as related to Transformers that led to natural language understanding breakthrough. But within it, is this incredible thing called large language models, which embeds human knowledge because it's been trained and so much data. And we recently announced Megatron 530B. And it was a collaboration we did with Microsoft, the foundation of – I think they call it Turing. And this language model and others like it, like open AI, GPD 3 are really transformative and they take an enormous amount of computation. However, the net result is a pre-trade model that is really quite remarkable. Now we're working with thousands of start-ups, large companies that are building who are using the public cloud. And so it's driving a lot of demand for us in the public cloud. I think we have now 10,000 AI inception startups that are working with us and using NVIDIA AI, whether it's on-prem or in the cloud, it saves money, because the computation time is significantly reduced. The quality of service is a lot better and they could do greater things. And so that's driving AI in the cloud. And so all of these different factors, whether it's just the industrial recognition of the importance of AI, the transformative nature of these new AI models recommender systems, large language models, conversational AI. The thousands of companies around the world that are using NVIDIA AI in the cloud -- driving public cloud demand, all of these things are driving our Data Center growth. And so we expect to see Data Center demand remain strong.

Tim Arcuri:

Thank you very much. I had a question about this $500 million impact for July and whether it's more supply related or demand related. And that's because most others in semis are sort of setting this China stuff, in particular, is more of a logistics issues, so more of a supply issues, but the language Colette you were using in your commentary side of lower sell-through in gaming and sort the absence of sales in Russia, to me that sounds a little more demand which would make sense in the context of this new freeze on hiring that you have. So, I ask because if it's supply related, then you could argue that it's not perishable and really just timing. But if demand related that might never come back and it to be the beginning of a falling night. So, I wonder if you can sort of walk through that for me? Thanks.

Operator:

Next, we'll go to Ambrish Srivastava with BMO. Your line is now open.

Jensen Huang:

Yes. Ambrish, thanks for the question. On first principles, it should be the case that our visibility of Data Centers is vastly better, vastly better than a couple of years ago. And the reason for that is several. One, if you recall a couple two, three years ago, deep learning and AI was starting to accelerate in the most computer science deep companies in the world with CSPs and hyperscalers. And -- but just about everywhere else, it was still quite nascent. And there was a couple of reasons for that. Obviously, the understanding of the technology is not as pervasive at the time. The type of industrial use cases for artificial intelligence requires labeling of data that's really quite difficult. And then now with Transformers, you have unsupervised learning and other techniques, zero-shot learning that allows us to do all kinds of interesting things without having to have human-labeled data. We even have synthetic generated data with Omniverse that helps customers do data generation without having to label data, which is either too costly or, quite frankly, oftentimes impossible. And so now, the knowledge and the technology has evolved to a place that most of the industries could use artificial intelligence at a fairly effective way and in many industries rather transformative. And so I think, number one, we went from clouds and hyperscalers to all of industries. Second, we went from training-focused to inference. Most people thought that inference was going to be easy. It turns out the inference is by far the harder. And the reason for that is because there are so many different models and there are so many different use cases and so many quality of service requirements, and you want to run these inference models in a small of a footprint as you can. And so when you scale out, the number of users that use the service is really quite high. So using acceleration, using NVIDIA's platform, we could inference any model from computer vision to speech to chemistry to biology, you name it. And we do it so quickly and so fast that the cost is very low. And so the more acceleration you do, the more money you will save. And that, I think, that wisdom is absolutely true. And so the second dimension is training to inference. The third dimension is that we now have so many different types of configurations of systems that we can go from high-performance computing systems all the way to cloud to on-prem to edge. And then the final concept is really this industrial deployment now of AI that's causing us to be able to in just about every industry, find growth. And so as you know, our cloud and hyperscalers are growing very, very quickly. However, the vertical part, vertical industries, which is financial services and retail and telco and all of those vertical industries have also grown very, very nicely. And so, in all of those different dimensions, our visibility should be a lot better. And then starting a couple of years ago, adding the Mellanox portfolio to our company, we're able to provide a lot more solution-oriented end-to-end platform solutions for companies that don't have the skills and don't have the technical depth to be able to stand up these sophisticated systems. And so, our networking business is growing very, very nicely as well.

Harlan Sur:

Hi, good afternoon. Thanks for let me ask the question. I just want to maybe just ask this question a little bit more directly. So, it's good to see the team being able to drive – navigate the dynamic supply chain environment, right? You look strong sequential growth in data center in April, here in the July quarter, even with some demand impact from Russia, right? And so, as we think about the second half of the year, cloud spending is strong, and it's actually, I think, accelerating. You're getting ready to ramp H100 later in the year. Mellanox, I think, is getting more supply as you move through the year. And in general, I think previously, you guys were anticipating sequential supply and revenue growth for the business through this entire year. I understand the uncertainty around gaming, but does the team expect continued sequential growth in data center through the remainder of the year?

Operator:

Next, we'll go to Chris Caso with Raymond James. Your line is open.

Colette Kress:

Yes. So let me start here, and I'll see if Jensen wants to add more of that. Our purchase obligations, as well as our prepaid have two major things to keep in mind. One, for the first time ever, we are prepaying to make sure that we have that supply and those commitments long term. And additionally, on our purchase obligations, many of them are for long lead time items that are a must for us to procure to make sure that we have the products coming to market. A good percentage of our purchase commitments is for our Data Center business, which you can imagine, are much larger systems, much more complex systems and those things that we are procuring to make sure we can feed the demand both in the upcoming quarters and further. Areas in terms of where we are still a little bit supply constrained are networking. Our demand is quite strong. We've been improving it each time. But yes, we still have demand -- excuse me, supply concerns with networking still. Is there others that you want to add on, Jensen?

Operator:

Our final question comes from Aaron Rakers with Wells Fargo. Your line is open.

Jensen Huang:

Several data points. We are just starting. We have just started shipping Orin in the first quarter of shipping production Orin. Orin is a robotics processor. It's designed for a software-defined robotic car or robotic pick and placer or a robotic mover, logistics mover. We've been designed into 35 car and trucks and robo taxi companies and more others, if you include logistics movers and last-mile delivery systems and farming equipment and the number of design wins for Orin is really quite fantastic. Orin is a revolutionary processor. And it's designed as a, if you will, a Data Center on a chip. And it is the first Data Center on a chip that is robotic, processes sensor information, it's safe, it has the ability to be rather resilient as confidential computing. It is designed to be secure, designed to be all those things because these data centers are going to be everywhere. And so Orin is really a technological marvels in production. We experienced very likely the lowest auto quarter in some time for some time. And the reason for that is because over the next six years-or-so, we have $11 billion and counting of business that we've secured estimated. And so I think it's a fairly safe thing to say now that Orin and our autonomous vehicle and robotics business is going to be our next multibillion dollar business. It's on its way surely there. The robotics and autonomous systems and autonomous machines, whether they move or not move, but AI systems that are at the physical edge is surely going to be the next major computing segment. It is surely going to be the next major Data Center segment. We've been working in this area, as you know, for a decade. We have a fair amount of expertise in this area. And Orin is just one example of our work here. We have four pillars to our strategy for autonomous systems. Starting from the data processing and the AI training part of it, to train robotics AIs; second, to simulate robotics AIs, which is omniverse; third, to the memory of the robotics AI otherwise known as mapping; and then finally, the actual robotics application and the robotics processor in the system, and that's where Orin goes. But Orin is just one of our four pillars of our robotics strategy and the next wave of AI. And so I am really optimistic and really enthusiastic about the next phase of the computer industry's growth. And I think a lot of it is going to be at the edge. A lot of it's going to be about robotics.

Jensen Huang:

Thanks, everyone. The full impact and duration of the war in Ukraine and COVID lockdowns in China is difficult to predict. However, the impact of our technology and our market opportunities remain unchanged. The effectiveness of deep learning AI continues to stand. The transformer model, which led to the natural language understanding breakthroughs is being advanced to learn patterns with great spatial, sequential and temporal complexity. Researchers are creating transformer models that are revolutionizing applications from robotics to drug discovery. The effectiveness of deep learning AI is driving companies across industries to adopt NVIDIA for AI computing. We're focused on four major initiatives. First, ramping our next generation of AI infrastructure chips and platforms, Hopper GPU, BlueField DPU, NVLink, InfiniBand, Quantum InfiniBand, Spectrum Ethernet Networking. And all this to help customers build their AI factories and take advantage of new AI breakthroughs like transformers. Second, ramping our system and software industry partners to launch Grace, our first CPU. Third, ramping Orin, our new robotics processor and nearly 40 customers building cars, robo taxis, trucks, delivery robots, logistics robots, farming robots to medical instruments. And fourth, with our software platforms, adding new value to our ecosystem with NVIDIA AI and NVIDIA Omniverse and expanding into new markets with new CUDA acceleration libraries. These initiatives will greatly advance AI. And while continuing to extend this most impactful technology of our time to scientists in every field and companies in every industry. We look forward to updating you on our progress next quarter. Thank you.

Operator:

Good afternoon. My name is David, and I will be your conference operator today. At this time, I'd like to welcome everyone to NVIDIA's Fourth Quarter Earnings Call. Today's conference is being recorded. All lines have been placed on mute to prevent any background noise. After the speakers’ remarks, there will be a question-and-answer session. [Operator Instructions]. Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. We had an excellent quarter with revenue up 53% year-on-year to $7.6 billion. We set records for total revenue as well as for Gaming, Data Center and Professional Visualization. Full year revenue was a record $26.9 billion, up 61%, compounding the prior year's growth of 53%. Starting with Gaming. Revenue of $3.4 billion rose 6% sequentially and was up 37% from a year earlier. Fiscal year revenue of $12.5 billion was up 61%. Gaming has become the top entertainment medium that continues to show strong momentum. Just last month, Steam hit a record 28 million concurrent users, up 50% in two years. Record desktop revenue in the quarter was led by our growth in our GeForce RTX 30 Series products with continued strength in the high end. At CES, we announced the RTX 3050 GPO, which hit retail in late January, bringing NVIDIA RTX and AI technologies to more mainstream audiences. Laptop gaming revenue also set a record, driven by the ramp of the new GeForce RTX 3070 Ti and 3080 Ti GPUs, which were also announced at CES. These leverage our fourth-generation Max-Q technology to enable quiet thin and light gaming laptops. All in, we announced over 160 new laptop designs for NVIDIA and tier architecture RTX 30 Series GPUs. These include a number of studio systems targeting the tens of millions of creators driving the future of design, innovation and virtual world. In addition to supporting the new RTX 30 Series GPUs, Studio laptops future support for NVIDIA software with Omniverse, Tundus and Broadcast. Availability of our gaming products in the channel remains low. NVIDIA RTX ecosystem continues to expand with over 30 new RTX games and applications added this quarter, including blockbuster hit like Battlefield 2042, Grand Theft Auto, Call of Duty

Colette Kress:

Thanks, Jensen. Moving to Data Center. Record revenue of $3.3 billion grew 11% sequentially and 71% from a year earlier. Fiscal year revenue of $10.6 billion was up 58%. Data center growth in the quarter was once again led by our compute products on strong demand for NVIDIA AI. Hyperscale and cloud demand was outstanding, with revenue more than doubling year-on-year. Vertical Industries also posted strong double-digit year-on-year growth led by consumer Internet companies. The flagship NVIDIA A100 GPU continue to drive strong growth. Inference-focused revenue more than tripled year-on-year. Accelerating inference growth has been enabled by widespread adoption of our Triton and France server software, which helps customers deliver fast and scalable AI in production. Data center compute demand was driven by continued deployment of our Ampere architecture-based product for fast-growing AI workloads such as natural language processing and deep learning recommendation systems as well as cloud executing. For example, Block Inc., a global leader in payment, uses conversational AI in its Square Assistant to schedule appointments with customers. These AI models are trained on video GPUs in AWS and perform inference 10x faster on the AWS GP service and on our CPUs. Social media company Snap used NVIDIA GPUs and Merlin deep recommendator software to improve inference cost efficiency by 50% and decrease latency to 2x. For the third year in a row, industry benchmarks show that NVIDIA AI continues to lead the industry in performance. Along with partners like Microsoft Azure, NVIDIA such records in the latest benchmarks for AI training across 8 popular AI workloads, including computer vision, natural language processing, recommendation systems, reinforcement learning and detection. NVIDIA AI was the only platform to make submissions across all benchmarks and use cases, demonstrating versatility as well as our performance. The numbers show performance gains on our A100 GPUs of over 5x in just 2 months, thanks to continuous innovations across the full stack in AI algorithms, optimization tools and system software. Over the past 3 years, they saw performance gains of over 20x powered by advances we have made across our full stack offering GPUs, networks, systems and software. The leading performance of NVIDIA AI is sought after by some of the world's most technically advanced companies. Meta Platforms unveils its new AI supercomputer research, SuperCluster, with over 6,000 A100 GPUs moved to an NVIDIA -- Meta's early benchmarks showed its system can train large natural language processing models 3x faster and run computer vision jobs 20x faster than the prior system. In a second phase later this year, the system will expand to 16,000 GPUs that Meta believes will deliver 5x of mixed precision AI performance. In addition to performance at scale, Meta cited extreme reliability, security, privacy and flexibility to handle a wide range of AI models as its key criteria for the system. We continue to broaden the reach and ease the adoption of NVIDIA AI into vertical industries. Our ecosystem of NVIDIA-certified systems expanded with Cisco and Hitachi -- which joined Dell, HewlettPackard Enterprise, Insper, Lenovo and Supermicro, among other sever manufacturers. We released version 1.1 of our NVIDIA AI Enterprise software, allowing enterprises to accelerated annual workloads on VMware, on mainstream IT infrastructure as well. And we expanded the number of system integrators qualified for NVIDIA AI Enterprise. Forrester Research in its evaluation of Enterprise AI infrastructure providers recognized NVIDIA in the top category of leaders. An example of a partner that's helping to expand our reach into enterprise IT is Deloitte, a leading global consulting firm, which has built its center for AI computing on NVIDIA DGX Superpod. At CES, we extended our collaboration to AV development, leveraging our own robust AI infrastructure and Deloitte's team of 5,500 system integration developers and 2,000 data scientists to architect solutions for truly intelligent transportation. Our networking products posted strong sequential and year-over-year growth, driven by exceptional demand across use cases ranging from computing, supercomputing and enterprise to storage. adopters-led growth driven by adoption of our next-generation products and higher-speed deployments. While revenue was gated by supply, we anticipate improving capacity in coming quarters, which should allow us to serve with significant customer demands we're seeing. Across the board, we are excited about the traction we are seeing with our new software business models, including NVIDIA AI, NVIDIA Omniverse and NVIDIA DRIVE. We are still early in the software revenue ramp. Our pipelines are building as customers across the industry seek to accelerate their pace of adoption and innovation with NVIDIA. Now let me turn it back over to Jensen for some comments on Arm.

Colette Kress:

Thanks, Jensen. We're going to turn to our P&L and our outlook. For the discussion of the rest of the P&L, please refer to the CFO commentary published earlier today on our Investor Relations lean. Let me turn to the outlook for the first quarter of fiscal 2023. We expect sequential growth to be driven primarily by Data Center. Gaming will also contribute to growth. Revenue is expected to be $8.1 billion, plus or minus 2%. GAAP and non-GAAP gross margins are expected to be 65.2% and 67%, respectively, plus or minus 50 basis points. GAAP operating expenses are expected to be $3.55 billion, including the Arm write-off of $1.36 billion. Non-GAAP operating expenses are expected to be $1.6 billion. For the fiscal year, we expect to grow non-GAAP operating expenses at a similar percent as in fiscal 2022. GAAP and non-GAAP other operating, other income and expenses are both expected to be an expense of approximately $55 million, excluding gains and losses on nonaffiliated investments. Non-GAAP tax rate are expected to be 11% and 13%, plus or minus 1%, excluding discrete items. Capital expenditures are expected to be approximately $350 million to $400 million. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight upcoming events for the financial community. We will be attending the Morgan Stanley Technology, Media and Telecom Conference in person on March 7. We will also be hosting a virtual Investor Day on March 22, alongside the GPU Technology Conference. This will follow Jensen's opening keynote, which we invite you to tune into. Our earnings call to discuss the results for our first quarter of fiscal 2023 is scheduled for Wednesday, May 27. We will now open the call for questions. Operator, will you please poll for questions?

Toshiya Hari:

Great. Thank you so much for taking the question. Jensen and Colette, I wanted to ask about Data Center. Colette, based on your guidance, you're probably guiding Data Center growth on a year-over-year basis to accelerate into the April quarter. You talked about hyperscale cloud growing more than 2x and enterprise verticals growing strong double digits in the January quarter. Can you kind of speak to the drivers for April and perhaps speak to visibility into the second half of the fiscal year as well in Data Center? Thank you.

Colette Kress:

I'll start, and I'll turn it over to Jensen. For Q1, our guidance can include an acceleration of Data Center from where we left in terms of Q4. We will have growth across several of our market platforms within Q1, both Data Center, Gaming and probably a couple of others. But yes, there is expected to be accelerated growth as we move into Q1. I'll turn it over to Jensen to talk about the drivers that we see for the quarter and also for the full year.

Operator:

Next, we'll go to C.J. Muse with Evercore ISI.

Colette Kress:

Thanks, C.J., for the question. I'll start on the data center supply. As we discussed last quarter and discussed today, we still have some supply constraints across some of our businesses. Networking in the Data Center business has been supply constrained. We're improving every single day. And we do expect to improve supply each quarter as we enter into fiscal year '23 here. So that is probably the key area within our Data Center. But from time to time, there can be other focused on do -- so I'll turn the rest of the question to Jensen in terms of how about the rest of the year as well.

Operator:

Next, we'll go to Joe Moore with Morgan Stanley.

Jensen Huang:

Yes. Thanks, Joe. We have a multiple-arm projects ongoing in the company from - connected from devices to robotics processors such as the new Orin that's going into autonomous vehicles and robotic systems and industrial automation, robotics and such. Orin is doing incredibly well. It started production. And as we mentioned earlier, it's going to drive an inflection point starting in Q2, but accelerating through Q3 and the several years after as we ramp into all of the electric cars and all of the robotic applications and robotaxis and such. We also have Arm projects with the CPU that you mentioned, Grace. We have Grace, and we surely have the follow-ons to Grace, and you could expect us to do a lot of CPU developments around the Arm architecture. One of the things that's really evolved nicely over the last couple of years is the success that Arm has seen in hyperscalers and data centers. And it's really accelerated and motivated them to accelerate the development of higher-end CPUs. And so you're going to see a lot of exciting CPUs coming from us. And Grace is just the first example. You're going to see a whole bunch of them beyond that. But our strategy is accelerated computing. That's ultimately what we do for a living. We, as you know, love it where there's any CPUs. If it's an x86 from any vendor. So long as we have a CPU, we could connect NVIDIA's platform to it and accelerate it for artificial intelligence or computer graphics, robotics and such. And so we love to see the expansion of CPU footprints, and we're just thrilled that Arm is now growing into robotics and autonomous vehicles and cloud computing and supercomputing and in all these different applications, and we intend to bring the full spectrum of NVIDIA's accelerated computing platform to NVIDIA Arm CPUs.

John Pitzer:

Just on the inventory purchase obligations, I think this was the fourth quarter in a row where you've seen greater than 30% sequential growth and is the first quarter where that number is now eclipsing kind of your quarterly revenue guidance. And so I guess I'm trying to figure out to what extent is this just a reflection of how tight things are across the semi industry? To what extent is this the poker tale of kind of how bullish you are on future demand? And relative to your commentary, that supply starts to get better throughout the year, should we expect that number to start to level off? Or as the mix moves more to data center and more to longer cycle times, more complicated devices should that number continue to grow?

Operator:

Next, we'll go to Tim Arcuri with UBS.

Jensen Huang:

Yes. NVIDIA is a software-driven business. Accelerated computing is a software-driven business. It starts from recognizing what domain of applications we want to accelerate and can accelerate and then building an entire stack from the processor to the system to the system software, the acceleration engines and potentially even the applications itself, like the software that we were mentioning earlier, NVIDIA DRIVE, NVIDIA AI and NVIDIA Omniverse. These are applications that sit on top of system software and are really valuable to the marketplace. The way to think about our software licensing -- so we've always been a software-driven business. But for the very first time, we have packaged licensable software on -- available to customers. The way that we license software for NVIDIA AI Enterprise is per node of server. There's some 20 million, 25 million servers that are installed in the world today in enterprises, not including clouds. We believe that every single server in the future will be running AI software. And we would like to offer an engine that enables enterprises to be able to use the most advanced the most trusted, the most utilized AI engine in the world. And so that is essentially the target market, if you will, for NVIDIA AI. The NVIDIA Omniverse is targeting -- is designed for creators contributing content to a virtual world and connect it to robots that are contributing to content in a virtual world. And so it's based on connections. There are 40 million designers and creators around the world. There are going to be hundreds of millions of robots. Every single car will essentially be a robot someday. And those are connections that will be connected into a digital twin system like Omniverse. And those are -- so the Omniverse business model is per connection per year. And in the case of NVIDIA DRIVE, we share the economics of the software that we deliver, if it's AB software or parking software or cabin-based AI software, whatever the licensing is or whatever the service, if it's an upfront license, we share the economics of that. If it's a monthly service subscription, we share the economics of that. But basically, for the cars that we are part of, that we're developing, the end-to-end service, we will get the benefits of the economics of that for the entire life of the fleet of the car. And so you could imagine, with 10 million cars, with a modern car lifetimes of 10 to 20 years, the economics and the market, the installed opportunity is quite high. And so our business opportunity is based on those factors. But our software business really, really started several years ago with virtual GPUs, but this year was when we really stepped it up and offered for the very first time NVIDIA AI Enterprise, Omniverse and DRIVE. And so I watch the spot, I think this is going to be a very significant business opportunity for us, and we look forward to reporting on it.

Vivek Arya:

Jensen, in the past, you mentioned about 10% or so adoption rate for AI among your customer base. I was hoping you would quantify where we are in that adoption curve that you tend to differentiate between the adoption differences between your hyperscale and enterprise customers? And then kind of related to that, is there an inorganic element to your growth now that you have over $20 billion of cash on the balance sheet? How are you planning to deploy that to kind of accelerate your growth also?

Operator:

And next, we'll go to Aaron Rakers with Wells Fargo.

Jensen Huang:

Thanks, Michael. Let's see. We -- RTX is an unqualified home run. RTX completely reinvented modern computer graphics. It made -- it brought forward ray tracing about a decade earlier than anybody thought possible. The combination of RTX with artificial intelligence, which enabled this technology we call DLSS, is able to not only do a ton more computation using our processors, but also engage the powerful Tensor for processors that we have in our GPUs to generate images, beautiful images. RTX is being adopted by just every game developer on the planet now. It's being adopted by just about every design tool on the planet now. And if not for RTX, Omniverse wouldn't be possible. We wouldn't be able to do physically based path tracing and simulate sensors like radars and LiDARs and ultrasonics and of course, cameras and simulate these cameras physically and still be able to deliver the type of performance that we deliver. And so RTX was a game changer for the industry. It reset modern computer graphics. And it was an enabler for us to build an entire new platform from Omniverse. We're about, I think, about 1/3 of the way through upgrading an installed base that is growing. You know that video games is now the world's largest gaming genre. And Steam over the last 2 years has grown by 50%. The number of concurrent players on Steam has grown tremendously. And in just the last couple of years, a brand-new game store from Epic came on, and it's already a multi-hundred million dollar business. I think it's close to $1 billion that they're doing incredibly well. I'm so happy to see it. And so the overall gaming market is growing and it's growing quite nicely. But the thing that, in addition to resetting computer graphics for our entire installed base, the growing of our installed base because gaming is growing. There are a couple of other growth dynamics that are associated with GeForce and RTX that's really quite brand new. One of them is hybrid work. This is a permanent condition. And we now are seeing across the board people who are designers and creators now have to set up essentially a new workstation or new home workstation design studio so that they could do their work at home. In addition, the creative economy, the digital economy, the creative economy is really, really doing fantastically because everything has to be done in 3D now. Print ads are done in 3D. So 2D print is done in 3D. Video is done in 3D. In live video broadcast video, the millions of influencers now augment their broadcast with rich augmented reality and 3D graphics. And so 3D graphics is now not just for video games and 3D content, it's actually used now for all forms of digital content creation. And so RTX has all of these different drivers working behind it, and we're definitely in the early innings of RTX.

Stacy Rasgon:

So you said that the growth in the next quarter is about $450 million, give or take, driven by Data Center. Can you give us some feeling for how that growth is being driven by units versus pricing versus mix and how those drivers might differ between Gaming and Data Center, if at all, for Colette?

Operator:

Next, we'll go to Harlan Sur with JPMorgan.

Jensen Huang:

Yes, absolutely. If not for the work that we did so closely with Mellanox, the scalability of DGX and DGX Super Pine and the research supercomputer that was just installed in Meta would just not be possible. The concepts of overlapping networking and compute, moving some -- moving computing into the fabric, into the network, the work that we're doing with Synchronoss and Precision Timing so that we could create Omniverse computers that obey the laws of physics and space time, these things are just simply not possible. The work that we're doing to bring cloud-native secure multi-tenancy to supercomputing wouldn't have been possible. The number of innovations, that are countless. And so I am so thrilled with the combination and so through what the work the Mellanox team are doing. We've accelerated road maps as a result of the combination that we could leverage a much larger base of chip design. BlueField's road map has been accelerated probably by about a year. The switch in -- the quantum switch and the spectrum switch, the SerDes are absolutely world-class, shared between Ethernet and InfiniBand and NVLink, absolutely the best servers in the world. And so the list of opportunities or the list of combination benefits is really quite countless. And so I'm super thrilled with that. With respect to networking growth, we should be growing. If we weren't supply constrained, we should be growing faster than overall CSP growth. And the reason for that -- the reason for that is twofold. The first is because the networking leadership position of Mellanox, Mellanox is highly heavyweight in the upper end of networking, where the adoption of higher-speed networks tends to move. And so it's sensible that as new data centers are built, the first preference is to install it with higher-speed networking than the last-generation networking. And Mellanox's networking technology is unambiguously world class. The second reason is because the areas where the overall NVIDIA is strong has to do with the areas that are growing quite fast, which related to artificial intelligence or cloud AI and such. And so those different applications are growing faster than the core. And so it would be sensible that we have the opportunity as we expand our supply base to continue to grow faster than CSPs overall.

Matthew Ramsay:

Yes. Jensen, I maybe wanted to expand on some of the things that you were just speaking about in your last answer with respect to the Data Center business. It's not often maybe ever that you have both x86 server vendors having new big platform upgrades in the same year, which will probably happen later this year. There's a lot going on there, PCIe, some CXL stuff. I wonder if you could talk a bit about your Data Center business broadly and what you feel might be memory and I/O constrained currently that these systems might unlock for you both in the cloud and enterprise side, but also in the DGX business.

Operator:

Our final question comes from the line of Raji Gill with Needham & Co.

Colette Kress:

Yes. Thanks for the question on gross margin and the long term. When we think about the long-term gross margin, we have incorporated software in many of our platforms even today, meaning our high-value platforms in data center or [indiscernible] of our business have really helped us with our gross margins to this point, and we've done a really solid job of managing that and the growth over the years. I believe these businesses will continue to be a growing opportunity for us, but now also with the ability to package up -- So as that scales with our Enterprise customers, in the Data center and with our already procured deals, a lot of work we've got a great opportunity in the future and [indiscernible] margin -- so we're going to work on that. We've set the stage for having been able to package it up to be able to sell it separately to create the business model, to create the partners that are helping us sell it. But yes, we do believe this will be a driver in the long term.

Jensen Huang:

Thanks, everyone. The tremendous demand for our computing platforms, NVIDIA RTX, NVIDIA HPC and NVIDIA AI drove a great quarter, capping a record year. Our work propels advances in AI, digital biology, climate sciences, gaming, creative design to autonomous vehicles and robotics and some of today's most impactful fields. Our open computing platform optimized across the full stack, architecture for data center scale is adopted by customers globally from cloud to core to edge and robotics. I am proud of the NVIDIA operations team as we make substantial strides in broadening our supply base to scale our company and better serve customer demand. And this year, we introduced new software business models with NVIDIA AI Enterprise, NVIDIA Omniverse and NVIDIA DRIVE. NVIDIA DRIVE is a full stack end-to-end platform that serves the industry with AV chips, data center infrastructure for AI and simulation, mapping and the autonomous driving application service. Our data center infrastructure is used by just about anybody building AVs, robotics robotaxis, shuttles and trucks. EV companies have selected our Orin chip across the world. And our partnership with Mercedes-Benz and Jaguar Land Rover has opened up a new software and services business model for millions of cars for the life of the fleet. NVIDIA Omniverse is a world simulation engine that connects simulated digital worlds to the physical world. Omniverse is a digital twin, a simulation of the physical world. The system can be a building a factory, a warehouse, a car, a fleet of cars, a robotic factory orchestrating a fleet of robots building cars that are themselves robotic. Today's Internet is 2D and AI is in the cloud. The next phase of Internet will be 3D and AI will be connected to the physical world. We created Omniverse to enable the next wave of AI where AI and robotics touches our world. Omniverse can sound like science fiction, but there are real-world use cases today. Hundreds of companies are evaluating Omniverse. We can't wait to share more of our progress at next month's GTC, learn about new chips, new computing platforms, new AI and robotic breakthroughs and the new frontiers of Omniverse. Hear from the technologists of Deloitte, Epic Games, Mercedes-Benz, Microsoft, Pfizer, Sony, Visa, Walt Disney, Zoom and more. This GTC promises to be our most exciting developers conference ever. We had quite a year, yet nothing makes me more proud than the incredible people who have made NVIDIA one of the best companies to work for and the company where they do their lives' work. We look forward to updating you on our progress next quarter. Thank you.

Operator:

Good afternoon. My name is Saydie, and I will be your conference operator today. At this time, I would like to welcome everyone to the NVIDIA's Third Quarter Earnings Call. [Operator Instructions] Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. Q3 was an outstanding quarter with revenue of $7.1 billion and year-on-year growth of 50%. We set records for total revenue as well as for Gaming, Data Center and Professional Visualization. Starting with Gaming. Revenue of $3.2 billion was up 5% sequentially and up 42% from a year earlier. Demand was strong across the board. While we continued to increase desktop GPU supply, we believe channel inventories remain low. Laptop GPUs also posted strong year-on-year growth, led by increased demand for high-end RTX laptops. NVIDIA RTX technology is driving our biggest ever refresh cycle with gamers, and continues to expand our base with creators. RTX introduced groundbreaking real-time ray tracing and AI-enabled super resolution capabilities, which are getting adopted at an accelerating pace. More than 200 games and applications now supports NVIDIA RTX, including 125 with NVIDIA DLSS. This quarter alone, 45 new games shipped with DLSS. And NVIDIA Reflex Latency Reducing Technology is in top esports titles, including Valorant, Fortnite, Apex Legends and Overwatch. In addition, the Reflex ecosystem continues to grow with Reflex technology now integrated in almost 50 gaming peripherals. NVIDIA studio for creators keeps expanding. Last month at the Adobe MAX Creativity Conference, Adobe announced two powerful AI features for Adobe Lightroom and the Lightroom Classic, accelerated by NVIDIA RTX GPUs. In addition, several of our partners launched new studio systems, including Microsoft, HP and ASUS. We estimate that a quarter of our installed base has adopted RTX GPUs. Looking ahead, we expect continued upgrades as well as growth from NVIDIA GeForce users, given rapidly expanding RTX support and the growing popularity of gaming, esports, content creation and streaming. Our GPUs are capable of crypto mining, but we don't have visibility into how much this impacts our overall GPU demand. In Q3, nearly all of our Ampere architecture gaming desktop GPU shipments were lite hash rate to help steer GeForce supply to gamers. Crypto mining processor revenue was $105 million, which is included in our OEM and other. Our cloud gaming service, GeForce Now, has two major achievements this quarter. First, Electronic Arts brought more of its hit games to the server. And second, we announced the new GeForce Now RTX 3080 membership tier, priced at less than $100 for six months. GeForce Now membership has more than doubled in this last year to over 14 million gamers for streaming content from 30 data centers in more than 80 countries. Moving to Pro Visualization. Q3 revenue of $577 million was up 11% sequentially and up 144% from the year ago quarter. The sequential rise was led by mobile workstations with desktop workstations also growing, as enterprises deployed systems to support hybrid work environment. Building on the strong initial ramp in Q2, Ampere architecture sales continue to grow, leading verticals, including media and entertainment, healthcare, public sector and automotive. Last week, we announced general availability of Omniverse Enterprise, a platform for simulating physically accurate 3D world and digital twins. Initial market reception to Omniverse has been incredible. Professionals at over 700 companies are evaluating the platform, including BMW, Ericsson, Lockheed Martin and Sony Pictures. More than 70,000 individual creators have downloaded Omniverse since the open beta launch in December. There are approximately 40 million 3D designers in the global market. Moving to Automotive. Q3 revenue of $135 million declined 11% sequentially and increased 8% from the year ago quarter. The sequential decline was primarily driven by AI cockpit revenue, which has negatively been impacted by automotive manufacturers supply constraints. We announced that self-driving truck start-up, Kodiak Robotics; auto maker, Lotus; autonomous bus manufacturers, QCraft; and EV start-up, WM Motor, have adopted NVIDIA DRIVE Orin platform for their next-generation vehicles. They join a large and rapidly growing list of companies adopting and developing on NVIDIA DRIVE, including auto OEMs, Tier 1 suppliers, NAVs, trucking companies, mobile taxis and software start-ups. Moving to Data Center. Record revenue of $2.9 billion grew 24% sequentially and 55% from the year ago quarter with record revenue across both hyperscale and vertical industries. Strong growth was led by hyperscale customers, fueled by continued rapid adoption of Ampere architecture Tensor Core GPUs for both internal and external workloads. Hyperscale compute revenue doubled year-on-year, driven by the scale out of natural language processing and recommendator models and cloud computing. Vertical industry growth was also strong, led by consumer Internet and broader cloud providers. For example, Barco Cloud deployed NVIDIA GPUs for its launch of AI services, such as tech analysis, speech recognition, computer vision and anomaly detection. We continue to achieve exceptional growth and influence, which again outpaced our overall Data Center growth. We have transitioned our lineup of infant-focused processes to the Ampere architecture, such as the A30 GPU. We also released the latest version of our Triton Inference Server software, enabling compute-intensive inference workloads such as large language models to scale across multiple GPUs and nodes with real-time performance. Over 25,000 companies worldwide use NVIDIA AI inference. A great new example is Microsoft Teams, which has nearly 250 million monthly active users. It uses NVIDIA AI to convert speech to text real time during video calls in 28 languages in a cost-effective way. We reached three milestones to help drive more mainstream enterprise adoption of NVIDIA AI. First, we announced the general availability of NVIDIA AI Enterprise, a comprehensive software suite with AI tools and frameworks that enables the hundreds of thousands of companies running NVIDIA, running vSphere to virtualize AI workloads on NVIDIA-certified systems. Second, VMware announced a future update to vSphere with Tanzu that is fully optimized for NVIDIA AI. When it's combined with NVIDIA AI enterprise, enterprises can efficiently manage cloud-native AI development and deployment on main stream data center servers and clouds with existing IT tools. And third, we expanded our launch cloud program globally with ethernet as our first digital infrastructure partner. NVIDIA LaunchPad is now available in nine locations worldwide, providing enterprises with immediate access to NVIDIA software and infrastructure to help them prototype and test data science and AI workloads. LaunchPad features NVIDIA-certified systems and NVIDIA DGX systems running the entire NVIDIA AI software stack. In networking, revenue was impacted as demand outstripped supply. We saw momentum toward higher speed and new generation products, including ConnectX-5 and 6. We announced the NVIDIA Quantum-2 400 gigabit per second end-to-end networking platform, consisting of the Quantum-2 switch the ConnectX-7 network adapter and the BlueField-3 DPU. The NVIDIA Quantum-two, which is available from a wide range of building infrastructure and system vendors around the world. Earlier this week, the latest top 500 list of supercomputers showed continued momentum for our full stack computing approach. NVIDIA's technologies accelerate over 70% of the systems Muslims on the including over 90% of all new systems and 23 of the top 25 most energy-efficient systems. Turning to GTC. Last week we hosted our GPU Technology Conference, which had over 270,000 registered attendees. Jensen's keynote has been viewed 25 million times over the past eight days. While our Spring GTC is focused on new chips and systems, this edition focused on software, demonstrating our full computing stack. Let me cover some of the highlights. Our vision for Omniverse came to life at GTC. We significantly expanded this ecosystem and announced new capabilities. Omniverse replication is an engine for producing data to train robots, replicating augment real-world data with massive, diverse and physically accurate synthetic data sets to both accelerate development of high-quality, high-performance AI across computing demand. NVIDIA Omniverse Avatar is our platform for generating interactive AI avatars. It connects several core NVIDIA SDKs including switch AI, computer vision, natural language understanding, recommendation engines and simulation. Applications including automated customer service, virtual collaboration and content solution. Replicator ad avatar joined several other announced features and capabilities for Omniverse, including AI, AR, VR and simulation-based technologies. We introduced 65 new and updated software development page, bringing our total to more than 150 serving industries from gaming and design to AI, cybersecurity, 5G and robotics. One of the STKs is our first four licensed AI model, NVIDIA Riva, for building conversational AI applications. Companies using Riva during the open beta include RingCentral for video conference live captioning and Pig An for customer service chatbots. NVIDIA Riva Enterprise will be commercially available early next year, for launch. We introduced the NVIDIA's NeMo Megatron optimized for training large language models on NVIDIA DGX SuperPOD infrastructure. This combination brings together production-ready, enterprise-grade hardware and software in both vertical industries, develop language and industry-specific dropbox, personal systems, content generation and summarization. Early adopters include SiDi, JD.com and VinBrain. We unveiled BlueField DOCA 1.2, the latest version of our GPU programming lender with new cybersecurity capabilities. DOCA is to our GPUs as CUDA is to our GPUs. It enables developers to build applications and services on top of our BlueFiled DOCAs. Our new capabilities make BlueField the ideal platform for the industry to build their own zero trust security platforms. The leading cybersecurity companies are working with us to provision their next-generation firewall service on BlueField, including Checkpoint, Juniper, Borgne, F5, Palo Alto Networks and VMware. And we released Clara Holoscan, a edge AI computing platform for medical instruments to improve decision-making tools in areas such as robo-assisted surgery, interventional radiology and radiation therapy planning. Other new or expanded SDKs or libraries unveiled at GTC include ReOpt for AI optimized logistics Quantum for quantum computing, Morpheus for cybersecurity, Modulus for physical-based machine learning and Crunet Numeric, a data center scale mass library to bring accelerated computing to the large and growing Python ecosystem. All in, NVIDIA's computing platform continues to expand as a broadening set of SDK enable more and more GPU-accelerated applications and industry use cases. CUDA has been downloaded 30 million times, and our developer ecosystem is now nearing 3 million strong. The applications they develop on top of our SDK and the cloud to edge computing platform are helping to transform multitrillion dollar industries from healthcare to transportation to mental services, manufacturing, logistics and virtual. In Automotive, we announced NVIDIA DRIVE Concierge and DRIVE Chauffeur, AI software platforms that enhance a vehicle's performance, features and safety. Live Concierge build on Omniverse Avatar functioned as an AI-based in-vehicle person-assistance, but enables automatic parking, summoning capabilities. It also enhanced safety by monitoring the driver throughout the duration of the drive. DRIVE Chauffeur offers autonomous capabilities, relieving the driver of constantly having to control the car. It will also perform address to address driving when combined with DRIVE Hyperion 8 platform. For robotics, we announced Jetson AGX Orin, the world's smallest, most powerful and energy-efficient AI supercomputer for robotics, autonomous mission and embedded computing at the Edge. Built on our Ampere architecture, Jetson AGX Orin provides 6 times the processing power of its predecessor and delivers 200 trillion operations per second, similar to a GPU-enabled server that fits into the palm of your hand. Jetson AGX Orin will be available in the first quarter of calendar 2022. Finally, we revealed plans to build Earth-2, the world's most powerful AI supercomputer dedicated to confronting climate change. The system would be the climate change counterpart to Cambridge-1, the U.K.'s most powerful AI supercomputer that we built for corporate research. Earth-2 furnishes all the technologies we've invented up to this moment. Let me discuss Arm. I'll provide you a brief update on our proposed acquisition of Arm. Arm with NVIDIA is a great opportunity for the industry and customers with NVIDIA's scale, capabilities and robust understanding of data center computing, acceleration and AI. We assessed Arm in expanding their reach into data center, IOT and PCs and advanced Arm's IP for decades to come. The combination of our companies can enhance competition in the industry as we work together on further building the world of AI. Regulators at the U.S. FTC have expressed concerns regarding the transaction and we are engaged in discussions with them regarding remedies to address those concerns. The transactions has been under review by China Antitrust Authority, pending the formal case initiation. Regulators in the U.K. and the EU have declined to approve the transaction in Phase 1 of their reviews on competition concerns. In the U.K., they have also voiced national security concerns. We have begun the Phase 2 process in the EU and U.K. jurisdictions. Despite these concerns and those raised by some Arm licensees, we continue to believe in the merits and the benefits of the acquisition to Arm, to its licensees and to the industry. We believe these concerns and those raised by some Arm licensees -- we continue to believe in the merits and benefits of the ongoing acquisition. Moving to the rest of the P&L. GAAP gross margin for the third quarter was up 260 basis points from a year earlier, primarily due to higher end mix within desktop, notebook, GeForce GPUs. The year-on-year increase also benefited from a reduced impact of acquisition-related costs. GAAP gross margin was up 40 basis points sequentially, driven by growth in our Data Center Ampere architecture products, which is particularly offset by mix in gaming. Non-gaming gross margin was up 150 basis points from a year earlier and up 30 basis points sequentially. Q3 GAAP EPS was $0.97, 83% from a year earlier. Non-GAAP EPS was $1.17, up 60% from a year ago, adjusting for our stock split. Q3 cash flow from operations was $1.5 billion, up from $1.3 billion a year earlier and down from $2.7 billion in the prior quarter. The year-on-year increase primarily reflects higher operating income, particularly offset by prepayment for long-term supply agreement. Let me turn to the outlook for the fourth quarter of fiscal 2022. We expect sequential growth to be driven by Data Center and Gaming, more than offsetting a decline in CMP. Revenue is expected to be $7.4 billion plus or minus 2%. GAAP and non-GAAP gross margins are expected to be 65.3% and 67%, respectively, plus or minus 50 basis points. GAAP and non-GAAP operating expenses are expected to be approximately $2.02 billion and $1.43 billion, respectively. GAAP and non-GAAP other income and expenses are both expected to be an expense of approximately $60 million, excluding gains and losses on non-affiliated investments. GAAP and non-GAAP tax rates are both expected to be 11%, plus or minus 1% excluding discrete items. Capital expenditures are expected to be approximately $250 million to $275 million. Further financial details are included in the CFO commentary. Other information is also available on our IR website. In closing, let me highlight upcoming events for the financial community. We will be attending the Credit Suisse 25th Annual Technology Conference in person on November 30th. We will also be at the Wells Fargo Fifth Annual TMT Summit virtually on December 1st, the UBS Global TMT Virtual Conference on December 6th, and the Deutsche Bank Virtual Auto Tech Conference on December 9th. Our earnings call to discuss our fourth quarter and fiscal year 2022 results is scheduled for Wednesday, February 16. With that, we will now open the call for questions. Operator, will you please poll for the questions.

Aaron Rakers:

Yes. Thanks for taking the question and congratulations on the results. I guess, I wanted to ask about Omniverse. Obviously, a lot of excitement around that. I guess the simple question is, Jensen, how do you define success in Omniverse as we look out over the next, let's call it, 12 months and how do we think about the subscription license opportunity for Omniverse. I know you've talked about $40 million total 3D designers, I think that actually doubled what you talked about back in August. So I'm just curious of how we at finance line should probably think about that opportunity materializing?

Operator:

For our next question, we have Mark Lipacis from Jefferies. Mark, your line is open.

Jensen Huang:

Yes. I really appreciate that. Chips are enablers, but chips don't create markets, software creates market. At this point, I explain over the years that accelerated computing is very different than general purpose computing and the reason for that is because you can't just write into compiler and compile quantum physics into a chip. I mean, it doesn't. You can't just compile Schrodinger's equation and have it distributed across multiple GPUs, multiple nodes and have a new SaaS. You just -- you can't do that for computer graphics, you can't do that for artificial intelligence, you can't do that for robotics, you can't do that for the most of the interesting applications in the world and because we really run out of steam with GPUs and that people are saying that not because it's not true, it is abundantly clear that the amount of instruction in parallel that you can squeeze out of a system is although not zero is incredibly hard, it's just incredibly hard. And there is another approach and we have been advocating accelerated computing for some time and now people really see the benefit of it, but it does requires a lot of work and yet the work basically says for every domain, for every application we have -- for every application in large domain that you'll use, you have to have a whole stack. And so whenever you want to open a new market by accelerating those applications or that domain of applications, you have to come up with a new stack and the new stack is hard, because you have to understand the application, you have to understand the algorithms, the mathematics, you have to understand computer science to distribute it across, to take something that was single threaded and make it multi-threaded and make something that we've done sequentially and make it process in parallel. You break everything, you break storage, you break networking, you break everything. And so it takes a fair amount of expertise and that's why we're saying that over the years, over the course of 30 years we have become a full-stack company, because we've been trying to solve this problem practically through decades. And so that's one. But the benefit, once you have the ability, then you can open new markets and we played a really large role in democratizing artificial intelligence and making it possible for anybody to be able to do it. Our greatest contribution is I hope when it's all said and done that we democratized scientific computing, so that researchers and scientists, computer scientists, data scientists, scientists of all kinds were able to get access to this incredibly powerful tool that we call computers to do advance research. And so every single year we're coming up with new stacks and we got a whole bunch of stacks we are working on and many of them are working on in plain sight, so that you see it coming, you just have to connect it together. One of the areas that we spoke about this time, of course, was Omniverse and you saw the pieces of it being built over time, and it took half a decade to start building Omniverse, but it built on a quarter century of work. In the case of the Omniverse Avatar, you could literally point to MERLIN, the recommender; Megatron, the language -- large language model; Riva, the speech AI, all of our computer vision AIs that have been demonstrating over the years, natural speech synthesis that we see every single year with I AM AI the opening credits, how we're using, developing an AI to be able to speak in a human way so that people feel more comfortable and more engaged with the AI. Face, eye tracking, Maxine and all of these technologies are connected together. They were all built in pieces, but we integrated it, we have the intentions of integrating it and to create what it's called Omniverse Avatar. And now you asked the question, how quickly will we deploy this? I believe Omniverse Avatar will be in drive-thrus and restaurants, fast food restaurants, check out with restaurants, in retail stores, all over the world within less than five years. And we're going to need it in all kinds of different applications, because there is such a great shortage of labor and there is such a wonderful way that you can now engage in Avatar and it could -- it doesn't make mistakes, it doesn't get tired and it's always on and we made so that it's cloud native and so when you saw the keynote, I hope you'd agree that the interaction is instantaneous and the conversational forum is so enjoyable. And so anyway, I think what you highlight is, one, accelerated computing is a full-stack challenge. Two, it takes software to open new markets. Chips can't open new markets. If you build another chip, you can steal somebody's share, but you can't open new market and it takes software to open new market. NVIDIA switch with software, and that's one of the reasons why we could integrate such large market opportunities. And then last with respect to Omniverse, I believe it's a near-term opportunity that we've be working on for some three, four, five years.

C.J. Muse:

Yes. Good afternoon. Thank you for taking the question. And I guess not an Omniverse question, but I guess, Jensen, I'd like your commitment that you will not use Omniverse to target the sell-side research industry. As my real question, can you speak to your Data Center visibility into 2022 and beyond? And within this outlook, can you talk to traditional cloud versus industry verticals and then perhaps emerging opportunities like Omniverse and others? Would love to get a sense of kind of what you're seeing today. And then as part of that, how you're planning to secure foundry and other supply to support that growth? Thank you.

Operator:

For our next question, we have Stacy Rasgon from Bernstein Research. Stacy, your line is open.

Colette Kress:

Okay. Thanks, Stacy, for the question. Let's first focus in terms of our guidance for Q4. Our statements that we made were, yes, about driven by revenue growth from Data Center and Gaming sequentially. We can probably expect our Data Center to grow faster than our Gaming, probably both in terms of percentage wise and in absolute dollars. We also expect our CMP product to decline quarter-on-quarter to very negligible levels in Q4. So I hope that gives you a color on Q4. Now in terms of next year, we'll certainly turn the corner into the new fiscal year. We certainly provide guidance one quarter out. We've given you some great discussions here about the opportunities in front of us, opportunities with the hyperscales, the opportunities with the verticals. Omniverse is a full stack opportunity in front of us. We are securing supply for next year, not just for the current year and Q4, to allow us to really grow into so much of this opportunity going forward. But at this time, we're going to wait until next year to provide guidance.

Operator:

For the next question we have Vivek Arya from BofA Securities. Vivek, your line is open.

Colette Kress:

Thanks for the question. So let's first start in terms of supply or supply purchase agreement. You have noted that we are discussing that we have made payments toward some of those commitments. Not only are we procuring for what we need in the quarter, what we need next year and again we are planning for growth next year, so we have been planning that supply purchases, we are also doing long-term supply purchases. These are areas of capacity agreements and/or many of our different suppliers. We made a payment within this quarter of approximately $1.6 billion out of total long-term capacity agreement of about $3.4 billion. So we still have more payments to make and we were likely continue to be purchasing longer term to support our growth that we are planning for many years to come.

Operator:

Your final question comes from the line of Timothy Arcuri from UBS. Timothy, your line is open.

Colette Kress:

Yes. So we have always been working on our gross margin and being able to absorb a lot of the cost changes along the way, architecture-to-architecture really. So that's always based into our gross margin. Our gross margins right now are largely stable. Our incremental revenue, for example, what we're expecting next quarter will likely align to our current gross margin levels that we finished in terms of Q3. Our largest driver always continues to be mix. We have a lot of different mix that has driven related to the high-end AI and RTX solutions, for example, and the software that is embedded in solutions have allowed us to increase our gross margin. As we look forward long-term, software if sold separately can be another driver of gross margin increases in the future. But cost changes, cost increases are -- generally been a part of our gross margin figures.

Jensen Huang:

Thank you. We had an outstanding quarter. Demand for NVIDIA AI is strong with hyperscalers and cloud services deploying at scale and enterprises broadening adoption. We now help more than 25,000 companies that are using NVIDIA AI. And with NVIDIA AI enterprise software suite, our collaboration with VMware and our collaboration with Equinix to place NVIDIA LaunchPad across the world, every enterprise has an easy arm length to NVIDIA AI. Gaming and Pro Vis are surging. RTX opportunity continues to expand with the growing market of gamers, creators, designers and now professionals building home workstations. We are working hard to increase supply for the overwhelming demand this holiday season. Last week, GTC showcased the expanding universe of NVIDIA accelerated computing. In combination with AI and Data Center scale computing, the model we pioneered is on the cusp of producing million X speed ups that will revolutionize many important fields; already AI and upcoming robotics, digital biology and what I hope climate science. GTC highlighted our full stack expertise in action, built on CUDA and our acceleration libraries in data processing, in simulation, graphics, artificial intelligence, market and domain specific software is needed to solve customer problems. We also showed how software opens new growth opportunities for us. But the chips are the enablers, but it's the software that opens new growth opportunities. NVIDIA has 150 SDKs now addressed in many of the world's largest end markets. One of the major themes of this GTC was Omniverse, our simulation platform for virtual worlds and digital twin. Our body of work and expertise in graphics, physics simulation, AI, robotics and full stack computing made Omniverse possible. At GTC, we showed how Omniverse is used to reinvent collaborative design, customer service avatars and video conferencing and digital twin to factories, processing plants and even entire cities. This is just the tip of the iceberg of what's to come. We look forward to updating you on our progress next quarter. Thank you.

Simona Jankowski:

Well, I think we just heard the closing remarks. Thank you so much for joining us. We look forward to seeing everybody at the conferences that we have planned over the next few months and I'm sure we'll talk before the end of next earnings. Thanks again, everybody.

Operator:

Good afternoon. My name is Mel, and I will be your conference operator today. At this time, I would like to welcome everyone to NVIDIA's Second Quarter Earnings Call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question and answer session. [Operator Instructions] Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. Q2 was another strong quarter with revenue of $6.5 billion and year-on-year growth of 68%. We set records for total revenue, as well as for gaming, data center, and professional visualization. Starting with gaming, revenue was 3.1 billion, was up 11% sequentially, and up 85% from a year earlier. Demand remained exceptionally strong, outpacing supply. We are now four quarters into the Ampere architecture product cycle for gaming, and it continues to be our best effort. At COMPUTEX in June, we announced two powerful new GPUs for gamers and creators, the GeForce, RTX 3080 Ti, and RTX 3070 Ti, delivering 50% faster performance than their prior-generation with acclaimed features, such as real-time retracing, NVIDIA DLSS, AI, rendering, Reflex, and broadcast. Laptop demand was also very strong. OEMs adopted Ampere architecture GPU s in a record number of designs. From the top of the line gaming laptops to those -- to mainstream price points as low as $799 that brings the power of GeForce GPUs to gamers, students, and creators on the go. Ampere architecture-powered laptops feature our third generation Max-Q power optimization technology that enables ultrathin designs, such as the new Alienware X15, the world's most powerful sub-16-millimeter gaming laptop. NVIDIA RTX technology has reset computer graphics and spurred our biggest ever refresh cycle. Ampere's been our fastest ramping gaming GPU architecture on Steam. And the combination of Turing and Ampere RTX GPUs has only upgraded about 20% of our installed base. 80% have yet to upgrade to RTX. And the audience for global eSports will soon approach 0.5 billion people, while the number of those who live stream games is expected to reach over 700 million. The number of PC gamers on Steam is up almost 20% over the past year. More than 60 RTX games now support NVIDIA's RTX ray tracing or DLSS, including today's biggest game franchises, such as Minecraft, Fortnite, and Cyberpunk. New RTX games this quarter includes Red Dead Redemption 2, one of the top-rated games of all time, popular titles like Rainbow Six Siege and rushed, and Minecraft RTX in China with over 400 million players. To competitive gamers, NVIDIA Reflex, which includes latency, is now supported by 20 games. Let me say a few words on crypto currency mining. In an effort to address the needs of miners and direct GeForce to gamers, we increased the supply of Crypto currency Mining Processors, or CMP, and introduced low hash rate GeForce GPUs with limited Ethereum mining capability. Over 80% of our Ampere architecture base GeForce shipments in the quarter were low hash rate GPUs. The combination of crypto to gaming revenue is difficult to quantify. CMP revenue, which is recognized in OEM, was 266 million, lower than our original 400 million estimates on reduced mining profitability and we expect a minimal contribution from CMP going forward. GeForce NOW reached a new milestone this quarter, surpassing 1,000 PC games, more than any other cloud gaming service. The premium tier is available for a subscription of $10 per month, giving gamers access to RTX class performance, even on an underpowered PC, Mac, Chromebook, iOS, or Android device. Moving to pro visualization. Q2 revenue was a record 519 million, up 40% sequentially, and up 156% year-on-year. Strong sequential revenue growth was led by desktop workstations, driven by demand to outfit, design offices at home as remote work becomes the norm across industries. This is also the first big quarter of the Ampere architecture ramp for pro visualization. King verticals driving Q2 demand include automotive, public sector, and healthcare. At SIGGRAPH last week, we announced an expansion of NVIDIA Omniverse, our simulation and collaboration platform that provides the foundation of the Metaverse. Through new integrations with Blender, the world's leading open-source 3D animation tool, and Adobe, we're opening the Omniverse platform to millions of additional users. We are also collaborating with Apple and Pixar to bring advanced physics capabilities to Pixar's Universal Scene Description framework, embracing open standards to provide 3D workflows to billions of devices. Omniverse enterprise software is in the early access stage and will be generally available later this year on a subscription basis from NVIDIA's partners, including Dell, HP, Lenovo, and many others. Over 500 Companies are evaluating Omniverse Enterprise, including BMW, Volvo, and Lockheed Martin. And more than 50,000 individual creators have downloaded Omniverse since it entered open beta in December. Moving to automotive. Our Q2 revenue was 152 million, down 1% sequentially, and up 3% year-on-year Sequential revenue declines in infotainment were largely offset by growth in self-driving. Looking further out, we have substantial design winds set to ramp that we expect will drive a major inflection in revenue in the coming years. This quarter, we announced several additional wins. Self-driving startup AutoX unveiled its latest autonomous driving platform for RoboTaxis powered by NVIDIA DRIVE. The performance and safety capabilities of the software-defined NVIDIA DRIVE platform have enabled AutoX to become one of the first companies in the world to provide full self-driving mobility services without the need for a safety driver. In autonomous trucking, DRIVE's ecosystem partner, plus, signed a deal with Amazon to provide at least 1,000 self-driving systems to Amazon's fleet of delivery vehicles. These systems are powered by NVIDIA DRIVE for high-performance, energy-efficient, and centralized AI compute. An autonomous trucking startup embarked is building on NVIDIA DRIVE. The system is being developed for trucks for four major OEMs, Freightliner, Navistar International, PACCAR, and Volvo, representing the vast majority of class 8 or largest size trucks in the U.S. The NVIDIA DRIVE platform is being rapidly adopted across the transportation industry from passenger-owned vehicles to robotaxi to trucking and delivery vehicles. We believe everything that moves will be autonomous someday. Moving to Data Center. Revenue of 2.4 billion grew 16% sequentially, and 35% from the year-ago quarter. The year-ago quarter, which was our first quarter to include Mellanox. Growth was driven by both hyperscale customers and vertical industries, each of which has record revenues. Our flagship A100 continuing to ramp across hyper-scale and cloud computing customers, with Microsoft Azure announcing general availability in June, following AWS and Google Cloud Platforms' general availability in prior quarters. Vertical industry demand was strong, with sequential growth led by financial services, supercomputing, and telecom customers. We also had exceptional growth in Inference, which reached a record more than doubling year-on-year. Revenue from Inference focussed processors includes the new A30 GPU, which provides four times the Inference performance of the T4. Customers are also turning to NVIDIA GPUs to take AI to production and shifting from CPUs to GPUs, driven by the stringent performance, latency, and cost requirements of deploying and scaling deep learning AI workloads, and NVIDIA Networking products posted solid results. We see momentum across regions driven by our technology of leadership with upgrades to high-speed products, such as ConnectX-6, as well as new customer wins across cloud service providers, enterprise, and high-performance computing. We extended our leadership in supercomputing. The latest top 500 niche shows that NVIDIA technologies power 342 of the world's top 500 supercomputers, including 70% of all new systems and eight of the top 10, to help companies harness the new industrial high-performance computing revolution. We deliver a turnkey AI Data Center solution with the NVIDIA DGX SuperPOD, the same technology that powers our new Cambridge-1 supercomputer in the UK and a number of others in the top 500. We expanded our AI software and subscription offerings, making it easier for enterprises to adopt AIs from the initial development stage through to deployment and operations. We announced NVIDIA Base Command, our software-as-a-service offering for operating and managing large-scale, multi-user, and multi-team AI development workloads on DGX SuperPOD. Base Command is the operating and management system software for distributed training customers. We also announced the general availability of NVIDIA Fleet Command, our managed edge AI software-as-a-service offering. Fleet Command helps Companies solve the problem of securely deploying and managing AI applications across thousands of remote locations, combining the efficiency and simplicity of central management with the cost performance and data sovereignty benefits of real-time processing at the edge. Early adopters of Fleet Command include some of the world's leading retail, manufacturing, and logistics companies and the specialty software Companies that work with them. The new NVIDIA Base Command and Fleet Command software and subscription offerings followed last quarter's announcements of the NVIDIA AI Enterprise software suite, which is early access with general availability expected soon. Our enterprise software strategy is supported by the NVIDIA certified system program with the server OEMs, which are bringing to market over 55 systems ready to run on NVIDIA's AI software out of the box to help enterprises simplify and accelerate their AI deployment. The NVIDIA ecosystem keeps getting stronger. NVIDIA Inception, our acceleration platform for AI startups, just surpassed 8,500 members with cumulative funding of over 60 billion and numbers in 90 countries. Inception is one of the largest AI startup ecosystems in the world. CUDA now has been downloaded 27 million times since it launched 15-years ago, with 7 million in the last year alone. TensorRT for inference has been downloaded nearly 2.5 million times across more than 27,000 Companies. And the total number of developers in the NVIDIA ecosystem now exceeds 2.6 million, up 4 times in the past 4 years. Let me give you a quick update on Arm. In nearly one year since we initially agreed to combine with Arm, we have gotten to know the Company, its business, and its people much better. We believe more than ever in the power of our combination, and the benefits it would deliver for Arm, for the UK, and its customers across the world in the era of AI. Arm has great potential. We love their business model and commit to keeping its open licensing approach. And with NVIDIA's scale and capabilities, Arm will make more IP, and sooner, for their mobile and embedded customers while expanding into Data Center, IOT, and other new markets. NVIDIA accelerates computing, which starts with the CPU. Whatever new markets are open with the CPU and our accelerated computing opportunities. We've announced accelerated platforms for Amazon Graviton, Ampere Computing, MediaTek, and more about expanding cloud computing, AI, cloud gaming, Supercomputing, Edge AI, to Chrome PCs. We plan to invest in the U.K. and we have with the Cambridge-1 supercomputer. And through ARM making U.K. a global center in science, technology, and AI. We are working through the regulatory process, although some Arm licensees have expressed concerns and objected to the transaction, and discussions with regulators are taking longer than initially thought. We are confident in the deal and that regulators should recognize the benefits of the acquisition to Arm, its licensees, and the industry. Moving to the rest of the P&L, the GAAP gross margin of 64.8% for the second quarter was up 600 basis points from the year earlier, reflecting the absence of certain Mellanox acquisition-related costs. GAAP gross margins were up 70 basis points sequentially, non-GAAP gross margins were 66.7% up 70 basis points from a year earlier and up 50 basis points sequentially, reflecting higher ASPs within the desktop, G4s, GPUs, our continued growth in high end and pure architecture products, partially offset by a mix shift within Data Center. Q2 GAAP EPS was $0.94 up 276% from a year earlier. Non-GAAP EPS was $1.04, up 89% from the year-earlier, adjusting for the 4 to 1 stock split effective this quarter. Q2 cash flow from operations was a record 2.7 billion. Let me turn to the outlook for the Third Quarter of Fiscal 2022. We expect another strong quarter with sequential growth driven largely by accelerating demand in Data Center. In addition, we expect sequential growth in each of our three other market platforms. Gaming demand is continuing to exceed supply as we expect channel inventories to remain below target levels as we exit Q3. The contribution of CMP to our revenue outlook is amongst. Revenue is expected to be 6.8 billion plus or minus 2%, GAAP and non-GAAP gross margins are expected to be 65.2% and 67% respectively, plus or minus 50 basis points. GAAP and non-GAAP operating expenses are expected to be approximately 1.96 billion and 1.37 billion, respectively. GAAP and non-GAAP other income and expenses are both expected to be an expense of approximately 60 million, excluding gains and losses on equity securities. GAAP and non-GAAP tax rates are supposed to be expected 11%, plus or minus 1% excluding discrete items. Capital expenditures are expected to be approximately 200 million to 225 million. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight upcoming events for the financial community. We will be attending the following virtual events. The BMO Technology Summit on August 24th, the New Street Big Ideas in Semiconductors Conference on September 9th, the Citi Global Tech Conference on September 13, the Piper Sandler Global Technology Conference on September 14th, and the Evercore ISI Auto Check and AI Forum on September 21st. Our earnings call to discuss the third-quarter results are scheduled for Wednesday, November 17th. We will now open the call for questions. Operator, would you please pull for questions?

Vivek Arya:

Thanks for taking my question. I actually had a near and longer-term question on the data center. I think near-term you mentioned the possibility of accelerating data center growth from the 35% rate. I was hoping if you could give us some more color around that confidence and visibility. And then longer-term, Jensen, we've seen a lot of announcements from NVIDIA about your Enterprise software opportunity. I honestly don't know how to model that. It sounds very promising, but how should we model it? What problem are you trying to solve? Is it cannibalizing demand you might have otherwise seen from your public Cloud customers, or is this incremental to growth? So just any guidance or any just insights into how to think about NVIDIA's Enterprise software opportunity longer-term? Thank you.

Operator:

Thank you. The next question comes from the line of Stacy Rasgon of Bernstein. Your line is now open. You may ask your question.

Colette Kress:

Yeah. Let me start and I'll let Jensen add a bit, Stacy, to your question. I guess we're providing the guidance for Q3 of 6.8 billion in revenue. Now excluding CMP, we expect our revenue to grow over 500 million sequentially. A lion's share of that sequential revenue increase will be coming from Data Center. We do expect gaming to be up slightly on a sequential basis but remember we are still supply constrained. Automotive and ProBiz are also expected to be up slightly quarter-over-quarter. And from the CMP perspective, we'll probably just have minimal amounts in Q3. So our Q3 results don't have seasonality with them for gaming and are really about the supply that we believe we can have for Q3. I'll see if Jensen wants to add any more colours.

Operator:

Thank you. The next question comes from the line of Matt Ramsay of Cowen. Your line is now open.

Colette Kress:

I'll start here. Thanks for the question on our overall mix as we go forward. First, our supply constraint in our gaming business is largely attributed to our desktop and notebook. That can mean a lot of different things from our components that are necessary to build so many of our products. But our mix is really important. Our mix as we are also seeing many of our gamers very interested in our higher-end, higher performance products. We will continue to see that as a driver about overall lifts both our revenue and can lift our overall gross margins. So there are quite a few different pieces into our supply that we have to think about, but we are going to try and make the best solutions for our gamers at this time.

C.J. Muse:

Yes, thank you. Good afternoon. I guess a follow-up question on the supply constraints. When do you think that they'll ease? And how should we think about gaming into the January quarter vis-a-vis typical seasonality, given -- I would assume you would continue to be supply constrained. Thank you.

Colette Kress:

Go ahead Jensen and I'll follow it up if there are some other things,

Operator:

Thank you. The next question comes from the line of Harlan Sur of JP Morgan. Your line is now open.

Jensen Huang:

Yeah, I really appreciate that question. Now, the heaviest solid growth quarter, and the Mellanox networking business have really grown incredibly. There are three dynamics happening all at the same time. The first is the transition that you're talking about. You know that the world's data center [Indiscernible] centers are users for computing costs just aggregated, which basically means with a single application is running on multiple servers at the same time. This is what makes it possible for them to scale up. The more users for an AI application or service, you just have to add more servers. And so the ease of scale-out that this aggregated computing provides also puts enormous pressure on the networking. And that Mellanox has the world's lowest latency and the highest bandwidth and performance networking on the planet. And so the ability to scale out and the ability to provide this aggregate in applications are really much, much better with Mellanox networking. So that's number one. Number two, almost every Company in the world has to be a high-performance computing Company now. You see that the cloud service providers, one after another, are building effective supercomputers. Historically was [Indiscernible] and supercomputing firms, the cloud-service providers have to build supercomputers themselves. And the reason for that is because artificial intelligence entombs gigantic models. The rate of growth of network sizes AI models buys this is doubled every two months. It's doubling not every year or 2 years, it's doubling every 2 months. And so you can imagine the size when I'm talking about trending AI models that are 100 trillion parameters large. The human brain has a 150 plus trillion synopsis, and so for nuance. And so that gives you a sense of the scale of AI models that people are developing. And so you're going to see supercomputers that are built out of Mellanox, InfiniBand, and their high-speed networking, along with NVIDIA GPU computing in more and more cloud service providers. You're also seeing it in enterprises or used in the discovery of [Indiscernible] there is our digital mileage and revolution going on as a computation statement. The large-scale computing that we're able to do now and AI, better understand biology and better understand chemistry, and bringing both of those skills into the field of Information Sciences. And so you're seeing large supercomputers growth in enterprises around the world as well. And so the second dynamic has to do with our incredibly great networking InfiniBand networking, which was the de facto standard in high-performance computing. And the third dynamic is the Data Center’s storing software. In order to orchestrate and run a data center, with just a few people essentially running their entire data center, hundreds of thousands of servers it is just one computer in front of you. That entire data center is software-defined. And the amount of software that goes into that software-defined data center running on today's GPUs is the networking stack, the storage stack, and now, because of zero trusts, the security stack. All of that is putting enormous pressure on the available computing capacity for applications, which is ultimately what data centers are designed to do. And so for the software-defined data center needs to have a place to take infrastructure software and accelerate it, to offload it, to accelerate it. And very important to isolate it from the application plan, so that intruders can't jump into the operating system with your all -- of your data center, establish of your data center. And so the answer to that is detailed, the ability to offload, accelerate, and isolate this -- the data center software infrastructure, and to fully up to all the -- your CPUs through to run what they're supposed to run, which is the application. Now, just about every data center in the world is moving towards a zero-trust model and Bluefield is just incredibly well-positioned. For these three dynamics, this aggregated computing, which needs really strong and fast networking, every Company needing high-performance computing, and then lastly, software decline in data centers going zero trusts. And so these are really important dynamics and I appreciate the opportunity to tell you all that. And you can just tell how super excited I'm about the prospects in the networking business and in the importance that they have in building modern data centers.

Aaron Rakers:

Yeah. Thanks for taking my question. I think you hit on a lot of my questions around the Data Center in that last. So maybe I will just ask the kind of on a P&L basis, one of the things that I see in the results, or more importantly, the guide is you're now collect guiding over a 67% gross margin, potentially. I'm curious as we move forward, how do you think about the incremental operating gross margin upside still from here, and how you're thinking about the operating merchant leverage for the Company from here through the P&L. Thank you.

Colette Kress:

Go forth.

Operator:

Thank you. We have the next question, which comes from the line of John Pitzer of Credit Suisse. Your line is open.

Jensen Huang:

Yes. There -- go ahead.

Jensen Huang:

There's the question about the strategy of how we're scaling GeForce supply to games. We moved incredibly fast this time with CMPs, and with our LHR settings for GeForce. And our entire strategy is about steering GeForce supply to the industry. And we have every reason to believe that because of the drive-in gaming, which is really a measure of gamers, the rate of growth of theme adoption of our GPUs, there's some evidence that that was successful. But there are somewhat reasons why it's just different this time. The first reason, of course, is that the LHR, which is new, and the speed at which we responded with CMPs, steered GeForce to fight the damage. The second is where at the very beginning of the Ampere and RTX graphics. As I mentioned earlier, RTX was a complete re-invention of computer graphics. Every evidence is that gamers are incredible -- and game developers are incredibly excited about ray tracing, this form of computer renderings. Graphics rendering is just dramatically more beautiful. And we're at the beginning of that cycle, and only 20% has been upgraded so far. We have 80% to go in a market that is already quite large and an installed base that's quite low but also growing. Last year, the gamers grew 20% and just [Indiscernible]. The third reserves that our demand is strong and our channel weak, and you can see what we're doing with the shortage of supply as quickly as we're shipping it. It's virtually all of the worlds. And then lastly, and we just have more growth drivers today because of RTX whenever. We have the biggest wave of NVIDIA laptops, just the laptops are our fastest-growing segment of computing and we have the largest wave of laptops coming. The demand for RTX and Workstation, whereas previously the Workstation market was a slow-growing market is now a fast-growing market and its achieved record. And after more than a decade of working on cloud graphics, our cloud graphics is in great demand. And so all of these segments are seeing high demand while we continue to supply them [Indiscernible] so I think the situations are very different and RTX is making a huge difference.

Chris Casper:

Thank you. Good evening. My question is about the split between the Hyperscale in the vertical customers in the Data Center business and the trends you see in each. I think in your prepared remarks, you said both would be up in the October quarter. But I'm interested to see if you're seeing any different trends there, particularly in the vertical business, as perhaps business conditions normalize and Companies return to the office, and they adjust their spending plans accordingly.

Jensen Huang:

There is a fundamental difference in the Hyperscale use of HPC or AI versus the industrial use of HPC Internet. In the world of hyperscalers and Internet service providers, they're making recommendations on movies, and songs, and articles, and search results, and so on, so forth. And the improvement that I can see that deep learning and artificial intelligence large recommender systems that can provide is really working for them. In the world of industry, the reason why artificial intelligence is transformative. Recognizing that most of the things that I just mentioned earlier, it's not moving on dynamics in the world's largest industries, whether it helps healthcare or in logistics or transportation or retail. The vast majority of the reasons why, and some of the physical sciences industries, whether it's energy or transportation, and also for healthcare. The simulation of physics, the simulation of the world, was not achievable using traditional first principle simulation approaches. But artificial intelligence or data-driven approaches have completely shaken that up and put it on its head. Some examples, whether it's fusing artificial intelligence so then you could feed up the simulation or the prediction of the approaching structure or the 3D structure approaching, which was recently achieved by a couple of very important networks, it's ground-breaking. And by understanding the approaching structure, 3D structure, we understand -- that we can better understand its function and how it would adapt to other approach and other [Indiscernible] And it's a fundamental step of the process in drug discovery and that has just taken a giant leap forward. In the areas of chronic science, it is now possible to consider using data-driven approaches to create models that overcome this -- not overcome but accelerate and make it possible for us to simulate much larger simulations of Multi-Physics Geometry aware simulations, which is basically climate science. These are really important fields of work that wouldn't have been possible for another decade at least. And just as we've made possible, using artificial intelligence, the realization of real-time ray tracing in every field of science, whether its climate simulation, energy discovery, drug discovery. We're starting to see that the industry recognizing that the fusion of the first principle simulation and data-driven artificial intelligence approaches, it's going to get a giant leap up. And that is a second dynamic. The other dynamic for the industry is for the very first time they can deploy AI model out to the edge to do a better job with agriculture, to do a better job with asset protection and warehouses, to do a better job with automating retail, AI is going to make it possible for all of these types of automation to finally be realized. And so the dynamics are all very different. That last one has to do with Edge AI, which was just made possible by putting AI right at the point of data and right at the point of action because you need to be low cost, you need to be high performance, and instantly responsive, and you can't afford to stream all of the data to the cloud all the time. And so each one of them has a partially different understanding.

Silgan Stein:

Great, thanks so much for taking my question. Jensen, I'm wondering if you can talk for a moment about Omniverse. This looks like really cool technology, but I tend to get very few questions from investors about it, but It looks to me like this could be a potentially very meaningful technology for you longer-term. Can you explain perhaps what capabilities and what markets this is going after? It looks like, perhaps this is going to position you very well in augmented and virtual reality, but maybe it's a different market or group of markets. It's a bit confusing to us, so if you could maybe help us understand it, I think we'd really appreciate it. Thank you.

Operator:

Thank you. I will now turn the call over back to Mr. Jensen Huang for closing remarks.

Operator:

Thank you. This concludes today's conference call. You may now disconnect.

Operator:

Good afternoon. My name is Sumitra, and I will be your conference operator today. At this time, I would like to welcome everyone to the NVIDIA's Financial Results Conference Call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer session. [Operator Instructions] Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. Q1 was exceptionally strong with revenue of $5.66 billion and year-on-year growth accelerating to 84%. We set a record in total revenue in gaming, data center and professional visualization driven by our best ever product lineups and structural tailwinds across our businesses. Starting with gaming. Revenue of $2.8 billion was up 11% sequentially and up 106% from a year earlier. This is the third consecutive quarter of accelerating year-on-year growth, beginning with the fall launch of our GeForce RTX 30 Series GPUs. Based on the Ampere GPU architecture, the 30 Series has been our most successful launch ever, driving incredible demand and setting records for both desktop and laptop GPU sales. Channel inventories are still leading, and we expect to remain supply constrained into the second half of the year. With our Ampere GPU architecture now ramping across the stack in both desktop and laptops, we expect the RTX upgrade cycle to kick into high gear as the vast majority of our GPU installed base needs to upgrade. Laptops continue to drive strong growth this quarter as we started ramping the Ampere GPU architecture across our lineup. Earlier this month, all major PC OEMs launch GeForce RTX 30 Series laptops based on the 3080, 39 -- 3070 and 3060 as part of their spring refresh. In addition, mainstream versions based on the 3050 and 3050 Ti will be available this summer, just in time for back-to-school starting at price points as low as 799. This is the largest ever wave of GeForce gaming laptops over 140 in total, as OEMs address the rising demand from gamers, creators and students for NVIDIA's powered laptops. The RTX 30 Series delivers our biggest generational leap in performance ever. It also features our second-generation ray tracing technology and frame rate boosting, AI-powered DLSS. The RTX is a reset for graphics with over 60 accelerated games. This quarter, we added many more, including Call of Duty, Modern Warfare, Crysis Remastered and Outriders. We also announced that DLSS is now available in Unreal Engine 4 and soon in the Unity game engine, enabling game developers to accelerate frame rates with minimal effort. The RTX 30 Series also offers NVIDIA Reflex, a new technology that reduces system latency. Reflex is emerging as a must-have feature for e-sports gamers who play competitive titles like Call of Duty

Timothy Arcuri:

Thanks a lot. Colette, I was wondering if you can double click a little more on the guidance. I know of the 600 to 650 in growth, you said 250 is coming from CMP and both gaming and data center will be up. Can we assume that they're up about equally, so you're getting about 200 roughly from each of those? And I guess second part of that is, within data center, I'm wondering can you speak to the networking piece. It sounds like maybe it was up a bit more modestly than it's been up the past few quarters. I'm just wondering what the outlook is there. Thanks.

Timothy Arcuri:

Thanks a lot, Colette.

Operator:

And your next question comes from C.J. Muse with Evercore ISI.

Colette Kress:

Let me take the first part of your question regarding our purchasing of inventory and what we're seeing in just both our purchase commitments and our inventory. The market has definitely changed to where long lead times are required to build out our data center products. So we're on a steady stream to both commit longer-term, so that we can make sure that we can serve our customers with the great lineup of products that we have. So yes, a good part of those purchase commitments is really about those long lead times, the components to create the full systems. I will turn the second part of the question over to Jensen.

Colette Kress:

Second part of the question was, what do we see in the second half as it relates to the lineup of enterprise. And we articulated in our pre-remarks regarding - that we see an acceleration. Thank you.

Operator:

And your next question comes from Aaron Rakers with Wells Fargo.

Colette Kress:

So, yes, we have definitely discussed in terms of the great opportunity that we have in front of us of folks moving to our ray traced GPUs. And we're in the early stages of that. We've had a strong cycle already, but still we probably have approximately 15% moving up a little bit from that at this time. So it's a great opportunity for us to continue to upgrade a good part of that install base, not only just with our desktop GPUs, but the RTX laptops are also a great driver of growth and upgrading folks to RTX.

Colette Kress:

Yes, please.

Operator:

And your next question comes from Vivek Arya with Bank of America Securities.

Jensen Huang:

Sure. Thanks a lot. First of all, it's hard to estimate exactly how much and where crypto mining is being done. However, we can only assume that the vast majority of it is contributed by professional miners, especially when the amount of mining increases tremendously like in-house [ph]. And so we created the CMP. And CMP and GeForce are not fungible. You could use GeForce for mining, but you can't use CMP for gaming. CMP is yields better and producing those doesn't take away from the supply of GeForce. And so it protects our GeForce supply for the gamers. And the question that you have is what happens when on the tail end of this? There's several things that we hope. And we learned a lot from the last time, but you never learn enough about this dynamic. What we hope is that that the CMPs will satisfy the miners at work will stay in mines, in the professional mines. And we're trying to produce a fair amount of them and we have secured a lot of demand for the CMPs and we will fulfill it. And what makes it different this time is several things. One, we're in the beginning of our RTX cycle, whereas Pascal was the last GTX. And now exactly was at the tail end of the GTX cycle, because the last GTX and it was the tail end of GTX cycle. We're at the very beginning of the RTX 30 cycle. And because we reinvented computer graphics, we reset the computer industry. And after 3 years, the entire graphics industry has followed. Every game developers need to do ray tracing, every content developer and every content tool has moved to ray tracing. And so if you lose ray tracing, these applications are so much better. And they simply run too slow on GTX's and so we're seeing a reset of the install base, if you will. And at a time when the gaming market is the largest ever, we've got this incredible install base of GeForce users. We've reinvented computer graphics and reset the install base and create an upgrade opportunity that's really exciting at a time when the market is the gaming market, the gaming industry is really large. And what's really exciting on top of that is that gaming is no longer just gaming. And it's infused into sports, e-sports. It's infused into art. It's infused into social. And so gaming is -- it has such a large cultural impact now, it's the largest form of entertainment. And I think that the experience we're going through is going to last a while. And so, one I hope that crypto will -- the CMP will steer our GeForce supply to gamers. We see strong demand and I expect to see strong demand for quite some time because of the dynamics that I described. And hopefully in the combination of those two, we'll see strong growth and through strong growth in our core gaming business through the year.

John Pitzer:

Yes. Good afternoon, guys. Thanks for let me ask the question. Jensen, I had two hopefully quick questions. First, I hearken back to the monitor you guys put out at couple of analyst days ago, the more you spend, the more you save. And you've always been very successful as you brought down the cost of doing something to really drive penetration growth. And so I'm curious with the NVIDIA Enterprise AI software stack, is there a sense that you can give us is how much that brings down the cost of deployment and AI inside the Enterprise? And do you think whether COVID lockdown related or cost related, there's pent up demand that this unlocks? And then my second question is just around government subsidies. A lot of talks out of Washington about subsidizing the chip industry, a lot of that goes towards building fabs domestically. But when I look at AI, I can't think of anything more important to maintain sort of leadership in relative to national security. How do we think about NVIDIA and kind of the impact that these government subsidies might have on either you or your customers or your business trends?

Operator:

And your final question comes from Stacy Rasgon with Bernstein.

Colette Kress:

Yes, Stacy, thanks for the question. So I first of all start with when we talked about our Q1 results. And when we're looking at Q1, we were really discussing a lot about what we expected between Q4 and Q1. Given what we knew was still high demand for gaming. We believed we would continue to grow between Q4 and Q1, which often we don't. And we absolutely have the strength and overall demand to grow. What that then lead was, again, continued growth from Q1 to Q2 as we are working hard to provide more supply for the strong demand that we say. We have talked about that we have additional supply coming. We expect to continue to grow as we move into the second half of the year as well for gaming. Now, we only guide one quarter at a time, but our plan is to take the supply, serve the overall gamers, work on building out the channel, as we know the channel is quite lean. And so yes, we do and still expect growth in the second half of the year, particularly when we see the lineup of games, the holiday overall coming, the back-to-school, all very important cycles for us. And there's a great opportunity to upgrade this, RTX install base. Now, in terms of data center, will work in terms of our guidance here. We have growth from Q1 to Q2 planned in our overall guidance. And we do see as things continue to open up a time to accelerate in the second half of the year for data center. We have, again a great lineup of products here. It couldn't be a better lineup now that we've also added the inferencing products and the host of overall applications that are using our software that we have. So this could be an opportunity as well to see that continued growth. We will work in terms of serving the supply that we need for both of these markets. But yes, we can see definitely growth in the second half of the year.

Jensen Huang:

Well, thank you. Thank you for joining us today. NVIDIA computing platform is accelerating. Launched at GTC, we are now ramping new platforms and initiatives. There are several that I mentioned. First, enabled by the fusion of NVIDIA RTX, NVIDIA AI and NVIDIA [indiscernible]. We built Omniverse, a platform for virtual collaboration and virtual worlds to enable tens of millions of artists and designers to create together in their own metaverse. Second, we lay the foundation to be a three check data center scale computing company with GPUs, DPUs and CPUs. Third, AI is the most powerful technology force of our time. We partner with cloud and consumer internet companies to scale out and commercialize AI powered services. And we're democratizing AI for every enterprise and every industry. With NVIDIA AGX certified systems, the NVIDIA Enterprise AI Suite pre-train models for conversational AI, language understanding, recommender systems and our broad partnerships across the IT industry, we are removing the barriers for every enterprise to access state-of-the-art AI. Four, the work of NVIDIA Clara in using AI to revolutionize genomics and biology is deeply impactful for the health care industry, and I look forward to telling you a lot more about this in the future. And fifth, the electric self driving and software defined car is coming. With NVIDIA DRIVE, we are partnering with the global transportation industry to reinvent the car architecture, reinvent mobility, reinvent driving and reinvent the business model of the industry. Transportation is going to be one of the world's largest technology industries. From gaming, metaverses cloud computing, AI, robotics, self driving cars, genomics, computational biology, NVIDIA is doing important work and innovating in the fastest growing markets today. As you can see, on top of our computing platforms that span PC, HPC, Cloud, Enterprise to Autonomous Edge, we've also transformed our business model beyond chips. NVIDIA vGPU, NVIDIA AI Enterprise, NVIDIA Fleet Command and NVIDIA Omniverse adds enterprise software license and subscription to our business model. And NVIDIA GeForce Now and NVIDIA DRIVE with Mercedes Benz as the lead partner, our end-to-end services on top of that. I want to thank all of the NVIDIA employees and partners for the amazing work you're doing. We look forward to updating you on our progress next quarter. Thank you.

Operator:

Good afternoon. My name is Mariama, and I will be your conference operator today. At this time, I would like to welcome everyone to NVIDIA's Financial Results Conference Call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer session. [Operator Instructions] Thank you. I will now turn the call over to Simona Jankowski, NVIDIA's Vice President of Investor Relations and Strategic Finance to begin the conference.

Colette Kress:

Thanks, Simona. Q4 was another record quarter with revenue exceeding $5 billion and year-on-year growth accelerating to 61%. Full-year revenue was also a record at $16.7 billion, up 53%. Our Gaming business have reached record revenue of $2.5 billion in Q4, up 10% sequentially, and up 67% from a year earlier. Full-year gaming revenue was a record at $7.8 billion, up 41%. Demand is incredible for our new GeForce RTX 30 Series products based on the NVIDIA Ampere GPU architecture. In early December, we launched the GeForce RTX 3060 Ti, which joined the previously launched RTX 3090, 3080, and 3070. The entire 30 Series lineup has been hard to keep in stock and we exited Q4 with channel inventories even lower than when we started. Although we are increasing supply, channel inventories will likely remain low throughout Q1. GeForce RTX 30 series graphics cards were a holiday sensation do not just do their amazing performance, but also to the rich features, including our second-generation RTX ray tracing technologies and DLSS, AI-powered performance accelerator, which massively boost frame rates in graphically demanding titles. Three dozen games now all support RTX, including the top Battle Royale game, Fortnite; the top role-playing game, Cyberpunk 2077; the top massively multiplayer online game, World of Warcraft; and the best-selling game of all time, Minecraft. RTX has clearly set the new standard in gaming. Building on this momentum at CES in January, we introduced a wave of Ampere architecture gaming products, including our biggest ever laptop launch powered by GeForce RTX 3060, 3070, and 3080 laptop GPUs and with our third-generation Max-Q technology. These new thin and lightweight gaming laptops increased performance and energy efficiency by up to 2x from the prior generation. RTX 3060 laptops start $999 and are faster than the previous generation laptops, which sold for $2,500. The incredible performance, design and price points of these new laptops will delight the growing universe of gamers and creators as well as students and professionals. The gaming laptop market has grown seven-fold in the past seven years and momentum is building. With top OEMs bringing to market a record 70-plus laptop models based on the GeForce RTX 30 series. GeForce laptops as a whole are the fastest-growing and one of the largest gaming platforms. Also at CES, we announced the GeForce RTX 3060 GPU, priced at $329, extending the 30 series desktop liner further into the mainstream. We expect strong demand when it launches this Friday as 60-class GPUs have traditionally been our most popular products. Starting with the 3060, we're taking an important step to maximize the supply of GeForce GPUs for gamers. Users are constantly discovering new applications for our powerful programmable GPUs and cryptocurrency mining is one of them. With rising Ethereum prices, there are indications that minors are behind GPUs. We would like GeForce GPUs to end up with gamers. So we have created a new special software drivers that will detect the Ethereum mining algorithm cutting in half the mining efficiency of the GeForce RTX 3060. We suspect the significant increase in the Ethereum network hash rate observed over the past few months was driven by a combination of previously installed mining capacities that was reactivated as well as new sales of GPUs and ASICs. Since our GPUs are sold to graphics card manufacturers and then onto distribution, we don't have the ability to accurately track or quantify their end use. Analyst estimate suggest that cryptomining contributed 100 million to 300 million to our Q4 revenue, a relatively small portion of our gaming revenue in Q4. Cryptocurrencies have recently started to be accepted by companies and financial institutions and show increased signs of staying power to address industrial Ethereum mining demand, last week we announced a new line of NVIDIA CMPs or cryptomining processors. Shipments will start in March. CMPs lack display outputs and have other optimizations that improve cryptomining power efficiency. CMP products will let us gain some visibility into the contribution of cryptomining to our overall revenue. For Q1, we estimate that CMP will contribute approximately 15 million. We plan to sell these products to industrial miners. We will quantify their contribution each quarter for transparency. Over the past year, it has become clear that we've entered a new era in which gaming is an integral part of global culture. The number of concurrent users on Steam has more than doubled since 2018 and continues to hit new records. In 2020 alone, more than 100 billion hours of gaming content was seen on YouTube and 0.5 billion people watched e-sports. Increasingly, we aren't just gaming. We're also watching sports, attending concerts, creating content and connecting with our friends in virtual environments. Additionally, we are excited about the new experiences like VR. Significantly more content is now available, including arguably the first VR killer app, Beat Saber, and there is now almost 2 million VR users on Steam. And with these powerful structural shifts, we expect our gaming business to remain on a robust growth trajectory. The GeForce RTX 30 Series GPUs have kicked off a powerful upgrade cycle and we estimate only around 15% of GeForce gamers on an RTX cost GPU, which is needed to experience the beautiful ray trace graphics of modern games. Moreover, the universe of gamers is rapidly expanding and the reach of GeForce has extended beyond gamers to some 45 million creators. In addition, gaming revenue continues to benefit from a favorable mix shift as gamers and creators keep moving to higher-end GPUs, we expect another great year for GeForce. Earlier this month, we celebrated the one year anniversary of the GeForce NOW cloud gaming platform, which is now over 6 million members strong. GeForce now offers 800 PCs from over 300 publishers, more than any other cloud gaming service including any of the most played free to play games, starting with support for windows PCs, Macs, and Android devices we added support in recent months to Chromebooks, iPhones and iPads. GFN has grown globally with more than 65 countries on our service and more – added regularly by our GeForce NOW alliance partners. Moving to Pro Vis, Q4 revenue was $307 million, up 30% sequentially and down 10% year-on-year and ahead of our expectations full-year revenue was $1.1 billion was down 13%. Strong sequential growth was driven primarily by a recovery in desktop workstations as some customers returned to the office and enterprises resumed purchases that had been deferred by the pandemic. Notebook GPUs grew sequentially to a record as enterprises continue to support remote workforce initiatives. Looking ahead, the reopening of businesses will benefit desktop workstations, but longer-term workforce trends will likely shift our mix to notebook GPUs and cloud offerings. Healthcare was a standout vertical in the quarter with significant orders from GE Siemens and Oxford Nanopore technologies, public sector and automotive also so strength. Omniverse are real time 3D collaboration and simulation platform is now an open beta. Over 500 creators and professionals have tested Omniverse through our early access program. Omniverse is one of our most important and exciting platforms. We are divided by its initial acceptance and look forward to sharing more details on its long-term growth opportunity in the coming months. Moving to Automotive; Q4 revenue was $145 million, up 16% sequentially and down 11% year-on-year. Full-year revenue of $536 million, declined 23%. Sequential growth was driven by continued recovering in the global automotive production volumes and growth in AI cockpit revenue. Year-on-year decline reflects the expected ramp down of legacy infotainment. Nvidia has emerged as the industry's leading end-to-end full stack technology provider for self-driving and AI enabled vehicles or in the SOC that drive self-driving platform is built on delivers an unrivaled 254 trillions of operations per second of performance on industry leading power efficiency, helping to revolutionize the transportation industry. Our technology leadership has driven a robust rapidly growing set of opportunities. We have great momentum with an expanding list of electric vehicle OEMs, including NIO, SAIC, Li Auto and Xpeng which are all using the Nvidia drive platform to power their next generation of vehicles. We look forward to growing with them as they continue to scale. Our software defined platforms is the only solution that spans from the data center for training deep neural net and running physically accurate simulations to a full stack in car solutions scaling for ADAS to Level 5 fully autonomous functionality. Autonomous vehicle companies are harnessing this technology. Zoox recently unveiled its Level 5 bi-directional robotaxi powered by Nvidia, and wide launched its next generation countless autonomous truck using NVIDIA Drive Orin. And earlier this year, Mercedes announced a 56-inch wide MBUX Hyperscreen powered by Nvidia AI cockpit technology. This will builds on our momentum with Mercedes first-generation MBUX system, which is now in 1.8 million cars. We are in the early innings of a significant opportunity. We have built a multi-billion dollar design win pipeline for our self-driving the AI cockpit solutions, which will drive a material inflection in revenue over the next few years. Our transformational partnership with Mercedes announced last June demonstrates the power of our evolving business model as we expand our addressable market and layer in software revenue. We are exceptionally well positioned to capitalize on the significant opportunity that lies ahead. Moving to data center; revenue was $1.9 billion, which exceeded our expectations was comparable to last quarter and up 97% from the year ago period, which did not include Mellanox. Data center compute revenue was up 45% year-on-year. Full-year data center revenue rose 125% to a record $6.7 billion, including almost 70% growth from data center compute. From a sequential perspective, the data center computes stronger than expected double-digit growth more than offset the anticipated decline in Mellanox revenue, which included a large non-reoccurring network sale to a single OEM in Q3. Compute growth was lead by vertical industries where OEM partners continued ramping up their 100 – day 100 base servers and our own DGX system sales were strong. Vertical industries were well over 50% of data center revenue across compute and networking with particular strength in super computing, financial services, higher education and consumer internet verticals. Additionally, hyperscale customers continued to deploy the A100 driving both sequential growth and exceptionally strong year-on-year growth in data center compute. The A100 has been adopted by all major cloud customers globally and is being deployed by hyperscale customers for internal workloads. Still we are in the early stages of adoption and expect continued growth this year. The ramp of the A100 has been smoother and accomplished by better visibility than prior generation. It's universal AI training and inference capabilities as well as support for wider set of applications and outstanding performance are driving five customer utilization; a clear sign of the A100's value. Turning to Mellanox. We are seeing continued strong traction and robust momentum across our customer set. Its revenue was up over 30% from Mellanox's Q4 revenue in calendar 2019 when it was still a standalone company. Year-on-year growth in the quarter was led by hyperscale and large consumer internet customers, which grew over 60% from last year with several contributing record revenues. Consistent with our outlook Mellanox had a sequential decline impacted by a non-reoccurring sales to a China OEM in Q3. We expect to return to sequential growth in Q1 driven by strong demand for our high speed networking products, including the ramp of ConnectX adapters with CSPs and all major server OEMs in their upcoming refresh. We also see strong momentum in high-performance computing with HDR InfiniBand products. For example, we won six of the seven supercomputers awarded over the past few months by Euro HPC. Starting next quarter, we will continue to provide color on networking as part of the data center market platform, but we will no longer break out Mellanox revenue separately. Looking forward, we are incredibly excited about the opportunities and data center. Accelerated computing is not only delivering super more loss, gains and performance but there's also an energy efficient and cost effective method of computing and virtually every industry is adopting technology with greater urgency as companies adapt to the new world of more distributed workers and customers. As industries embark on this journey, they are also increasingly focused on combating climate change. To that end, the A100 performed AI computations with one-twentieth the power consumption of CPU's. And powers are saline supercomputer, which is number one on the Green500 list of the world's most efficient supercomputers. Indeed Nvidia's powered machines recently captured 25 of the top 30 spots on the Green500 list. Accelerated computing is not only serving the exponential growth and demand for compute; it can also help bend the power consumption curve. With accelerated computing Nvidia is pioneering a [indiscernible] forward the computing industry. Before I moved to the P&L and outlook, let me give you an update on our proposed acquisition of Arm. In September, we announced plans to acquire Arm from SoftBank Group and a transaction that will create the premier computing company for the age of AI. At that time, we said it would take approximately 18 months to secure regulatory approvals in the U.S., the UK, the EU, China and other jurisdictions. Thorough reviews are typical with the deal of this size. This process is moving forward as expected. We are in constructive dialogue with the relevant authorities and are confident that regulators will see the benefits to the entire tech ecosystem. As we have said, this combination will spur competition together Arm and Nvidia will provide greater choice to the data center ecosystem, a compelling alternative CPU architecture for the market and further enhance Arms' offering in mobile and embedded. Our intention is to increase investment in Arm's existing roadmap, adding resources to stimulate growth in new markets. We love and intent to maintain Arm's open licensing model, a commitment guaranteed both by long-term legally binding contracts, as well as our own interest in ensuring this investment is a profitable one for us. We are on the cusp of new age in which AI fuels industries ranging from healthcare to scientific research to the environment. With this transaction, our vision is to boost Arm's potential, so it can thrive in this new era and grow into promising new markets. Moving to the rest of the P&L. Q4 GAAP gross margins were 63.1% and non-GAAP gross margins were 65.5%. GAAP gross margins declined year-on-year due to amortization of developed technology acquired from Mellanox, partially offset by product mix. The sequential increase was due to higher margins for gaming GPU and lower IP related costs, partially offset by lower margin mix in our data center portfolio. Non-GAAP gross margins increased by 10 basis points year-on-year and was flat sequentially in-line with our expectations. Q4 GAAP EPS was $2.31, up 51% from a year earlier. Non-GAAP EPS was $3.10, up 64% from a year ago. Q4 cash from operations was a record $2.07 billion. With that, let me turn to the outlook for the first quarter of fiscal 2022. Revenue is expected to be $5.3 billion plus or minus 2% with most of the sequential growth driven by gaming. GAAP and non-GAAP gross margins are expected to be 63.8% and 66% respectively plus or minus 50 basis points. GAAP and non-GAAP operating expenses are expected to be approximately $1.67 billion, and $1.2 billion respectively. For the full-year we expect to grow non-GAAP OPEX in the mid-20% range. GAAP and non-GAAP, other income and expenses are both expected to be an extent of approximately $50 million. GAAP and non-GAAP tax rates are both expected to be 10% plus or minus 1%, excluding discrete items. Capital expenditures are expected to be approximately $300 million to $325 million. Further financial details are included in the CFO commentary and other information on our IR website. In closing, let me highlight upcoming events for the financial community. We will be virtually attending to Raymond James Institutional Investors Conference on March 1st. The Morgan Stanley Technology, Media and Telecom Conference on March 3rd; and the Arete Virtual Semis Conference on March 3rd. In addition, we will be hosting a virtual Investor Day on Monday, April 12th following a live stream of Jensen's opening keynote at our GPU Technology Conference. Our earnings call to discuss our first quarter and full – our first quarter is scheduled for Wednesday, May 26. We will now open the call for questions. Operator, would you please poll for questions. Thank you.

C.J. Muse:

Good afternoon. Thank you for taking the question. I guess, Jensen, higher-level question for you on the enterprise side. You're now a couple of quarters into the ramp of A100.And curious if you could speak to whether you've seen any surprises here, any areas of specific strength worth calling out? And any changes to how you're thinking about the size of this opportunity?

Operator:

Your next question comes from the line of Vivek Arya with BofA Securities. Your line is open.

Colette Kress:

So thank you, Vivek. Your question regarding the guidance as we beat into Q1. We had indicated that, yes, a good percentage of our growth between Q4 and Q1 those come from gaming, but we also do expect data center to grow. Most of our sequential growth coming from gaming, but keep in mind, we also expect all of our market platforms will likely be able to grow quarter-over-quarter.

Operator:

Your next question comes from the line of Stacy Rasgon with Bernstein Research. Your line is open.

Jensen Huang:

Colette, you will take one and I'll take the one.

Jensen Huang:

Thanks, Colette. You asked the question about normal time, our company – at the company level where supply constraint, our demand is greater than our supply. And however for data center, so long as the customers more closely with us, and we do a good job planning between my companies, there should be a supply issue for data center. We just had to do a good job planning and we have direct relationships with each one of the royalty of fees and we have direct relationships with all the young ones. And we could do excellent planning between us. We shouldn't have a supplier. We shouldn't be supply constrained there. But at the company level, we're supply constrained. Demand is greater than supply. And yes, we have enough supply. Yes, we usually have enough supply to achieve that even the outlook. And we had that situation in Q4. We expect that situation in Q1. And we have enough supply to grow through the year. But supply against constraint and demand is really, really great. And so we just have to do a really good job planning. And meanwhile, one of the things that, that really came through for us is we have the world's best operations team. Our company is really, really have had an amazing operations team. We build the most complex products in the world, the most complex chips, the most complex packages, the most complex systems. And during Q4, they improved our cycle time and during Q1 we expecting them to improve our cycle time again. And we really are blessed to have such an amazing operations team. And so during these times it really comes in handy. But overall at the company level where we expected them to be greater than supply, we have enough supply to do better in the outlook. And we have enough supply to grow each quarter throughout the year.

Timothy Arcuri:

Hi, thanks. I had a question on crypto, I guess Jensen, I know that the CMP stuff and the software drivers stuff that you're doing for the 36. And that's going to help a lot, but I think there is like four or five the big currency was in a move or at least a moving or there on a faster move from a proof of work to proof of stake, which is going to be a lot less compute intensive. So I guess the question that I get a lot is how do you assess the degree to which that drives GPUs back into the secondary market? Is there any way to get treated, get kind of a handle on that? Thanks.

Operator:

Your next question comes from the line of John Pitzer with Credit Suisse. Your line is open.

Jensen Huang:

We're expecting – Mellanox was down this last quarter. And our compute business grew double digit and offset – more than offsets the decline in Mellanox. We expect Q1 to be a growth quarter to Mellanox and we expect this coming year to be quite an exciting year of growth for Mellanox. The business is growing and ethernet is growing for CSPs is growing in InfiniBand for high performance computing and the Switch – the Switches have grown. Switch business grew 50% year-over-year. And so, we're seeing really terrific growth there. One of the new initiatives and we're going to see success towards the second half because the number of adoptions, the number of engagements has grown as our BlueField DPUs. It's used for virtualization for hyperscalers. It's also used for security. As you know quite well the future of computing and cloud and it's multi-tenant cloud and there's no VPN front door to the cloud. You've got millions of people who are using every aspect of computing. So you need to have distributed firewalls and you can't have it just in one place. The intense focus of security across all of the data centers around the world is really creating a great condition for BlueField, which is really perfect then. And so, I expect our Mellanox networking business to grow very nicely this year. And we expect Q1 to be a great growth quarter for compute as well as Mellanox. The Killer, great driving application for AI are several – last year you're absolutely right that it with natural language understanding and the transformer model and what is the – what was the core of – and other versions like that really, really made it possible for us to enable all kinds of new applications. So you're going to see a natural language understanding do text completion, and it's going to be integrated – I think it was just announced today that it was going to be integrated into Microsoft Word, and we've been working with them on that for some time. And so there are some really exciting applications out there, but the new ones that came – that emerged recently are deep learning based conversational AI, where the ASR, the speech recognition as well as the speech synthesis are now based on deep learning, it wasn't before. And they were based on models that ran on CPUs, but now with these deep learning models, the accuracy is much, much higher and it has the ability to also mimic your voice and be a lot more natural. And so, the ability – these models are much more complex and much larger. The other big huge driver is the recommenders. This is something really worthwhile to take a look at is called deep learning recommender models, and recommenders have historically – whether it's for shopping and or personalizing websites are personalizing your store, recommending your basket, recommending your music. Historically, it's been – and use the traditional machine learning algorithms, but because of the accuracy and – just the extraordinary economic impact that comes from an incremental 1% in accuracy for most of the – mostly the world's large internet businesses people are moving very rapidly to deep learning based models. And these models are gigantic. They're utterly gigantic. And this is an area that is really driving high-performance computers. And – so we – I expect us to see a lot of momentum there. And the last one is the one that I just spoken out, which has to do with industrial 5G and edge, IoT type of applications for all of the different industries whether it's retail or logistics or transportation, agriculture or warehouses to factories. And so, we're going to see AI and robotics in a very large number of applications in industries and we're just seeing so much excitement there.

Aaron Rakers:

Yes. Thanks for taking the questions. I wanted to go back again on the data center business that you just mentioned, Jensen, that the BlueField-2 product poised to kind of ramp and materialize in the back half of the calendar year. How do you see that – is it an attach rate? I think there has been discussions in the past about – all servers could potentially over time incorporate this layer of acceleration. How quickly should we think about that ramp? And then the second question is, can you just at a high level talk about how CPU – how a CPU strategy you're thinking about that in the context of the broader data center market?

Operator:

Your next question comes from the line of Mark Lipacis with Jefferies. Your line is open.

Jensen Huang:

Yes, we're – I think we've done a great job on building out the platforms for several ecosystems around the world. And the domain that we do incredibly well out on – the domains that I have to do with accelerated computing, we pioneered this approach. And we brought it to high-performance computing at first and we accelerated scientific computing and we democratized supercomputing for all researchers, anybody who wants to have a supercomputer now can. And computing, it will simply not be the obstacle that somebody's discovery. We did the same for artificial intelligence. We did the same for visualization. We brought – we expanded the nature of gaining tremendously. Our GeForce today is the largest gaming platform. It’s the largest single largest body of computers that are used for gaming. And in each case, we expanded the market tremendously. We would like to do the same for data center scale computing, as it applies to virtualizing these applications, these applications are also in the process. They've historically required dedicated systems, but they're moving into a virtualized data center environment. And we are best at doing that. They run on our platform today. We have the ability to virtualize it and put it into the data center and make it remotely available. And so these applications, these domains are some of the most important domains in the world. And so we're in the process of getting them. By doing so and making our architecture available to CSPs and OEMs, we could create this accelerated computing platform available to everybody. And so that's we're seeing our journey doing them. First, creating an architecting this platform, and then putting it literally into every single data center in the world. But we would also like to the next step of our journey is there's the Phase 3 of AI and has to do about – it has to do with turning every end point into a data center, whether it's a 5G tower, a warehouse or retail store, a self-driving car, a self-driving truck, these are going to be – they're all going to be essentially autonomous data centers and/or they're going to run AI, but they're going to run a lot more. They're going to do security in real time. Its networking is going to be incredible. It's going to run software to 5G and GPO accelerating 5G, we call areal. And so these platforms are going to become data centers. There'll be secure. The software is protected and we can't tamper with. It if you tamper with it, of course won't run. And so the capability of these clouds will move all the way out to the edge. And we're in the best position to be able to do that. So I did the – in this new world of post close to Moore's law post in arts gaming in this new world where AI and software that writes software in this new world, where data centers are going to be literally everywhere and they're unprotected. There's no giant building with a whole bunch of people that secured. And in this new world where a software is going to enable this autonomous feature, I think we are a perfectly positioned for it.

Jensen Huang:

Thanks for joining us today. Q4 have the truly breakout year for Nvidia. The two biggest engines of our business gaming and data center posted powerful growth. Gaming has become the world's largest media and entertainment industry, and will grow to be much larger. And again gamers will create, will play, they'll learn, they'll connect. The medium of gaming can host any type of game and eventually evolve with the countless metaverses. So I'm for placed on for work. Gaming at simultaneously a great technology and a great business driver for our company. This year, we also closed our Mellanox acquisition and successfully united the amazing talent of our companies. Combined we possess deep expertise in all aspects of computing and networking to drive the architecture of modern data centers. Crowd computing and hyperscalers have transformed the data center into the new unit of computing. Chips on servers are just elements of the data center scale computers now. With our expertise in AI computing full stack accelerated computing, our deep network to computing expertise and cloud to edge platforms, Nvidia is helping to drive a great computer industry transformation. And our planned acquisition of Arm, the world's most popular and energy efficient CPU company will help position and video to lead in the age of AI. This year was extraordinary. The pandemic will pass, but the world has been changed forever. Technology adoption is accelerating across every industry. Companies and products need to be more remote and autonomous. This will drive data centers, AI and robotics. This underlies the accelerated adoption of Nvidia’s technology. The urgency to digitize, automate and accelerate innovation has never been higher. We are ready. We look forward to updating you on our progress next quarter. Thanks a lot.

Operator:

Good afternoon. My name is Jason, and I will be your conference operator today. At this time, I would like to welcome everyone to NVIDIA’s Third Quarter Financial Results Conference Call. All lines have been placed on mute to prevent any background noise. After the speakers’ remarks, there will be a question-and-answer session. [Operator Instructions] Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thank you, Simona. Q3 was another exceptional quarter with record revenue of $4.73 billion, up 57% year-on-year, up 22% sequentially and well above our outlook. Our new NVIDIA Ampere GPU architecture is ramping with excellent demand across our major market platforms. Q3 was also a landmark quarter, both for us and the industry as a whole. As we announced plans to acquire Arm from SoftBank for $40 billion, we are incredibly excited about the combined company’s opportunities and we are working through the regulatory approval process. For today, we will focus our remarks on our quarterly performance. Starting with gaming. Revenue was a record $2.27 billion, up 37% year-on-year, up 37% sequentially, and ahead of our high expectations. Driving strong growth was our new NVIDIA Ampere architecture-based GeForce RTX 30 series of gaming GPUs. The GeForce RTX 3070, 3080 and 3090 GPUs offer up to two times the performance and two times the power efficiency over the previous Turing-based generation. Our second generation NVIDIA RTX combines ray tracing and AI to deliver the greatest ever generational leap in performance. First announced on September 1st and ranging in price from $499 to $1,499, these GPUs have generated amazing reviews and overwhelming demand. PC World called them staggeringly powerful, while Newegg cited more traffic than Black Friday. Many of our retail and e-tail partners sold out instantly, the RTX 30 series drove our biggest ever launch. While we had anticipated strong demand, it exceeded even our bullish expectations. Given industry-wide capacity constraints and long cycle times, it may take a few more months for product availability to catch up with demand. In addition to the NVIDIA Ampere GPU architecture, we announced powerful new tools for gamers as well as for tens of millions of live streamers, broadcasters, eSports professionals, artists and creators. NVIDIA Reflex is a new technology that improves reaction time in games, reducing system latency by up to 58%. NVIDIA Reflex is being integrated into popular eSports games, such as Apex Legends, Call of Duty

John Pitzer:

Can you hear me?

John Pitzer:

Yes. Hey, guys. Congratulations on the solid results. Thank you for letting me ask the question. Just Colette, going back to your commentary around Mellanox, it seems like you’re guiding the January quarter to about $500 million, which means the core data center business is still growing nicely, call it, 6%, 7% sequentially. I’m just kind of curious, when you look at the core data center business, I know there’s not a direct correlation to server business. But, we’re clearly going through a cloud digestion in server and core vertical markets, enterprise, the servers are weak. When you look at your core data center business, do you feel as though that’s having an impact, and this is sort of the digestion that you saw kind of in late fiscal ‘20 -- sorry, fiscal ‘19 into ‘20, but you’re doing it still growing significantly year-over-year, or how would you characterize the macro backdrop?

Jensen Huang:

Yes. Colette captured it very well. The only thing I would add is that our inference initiative is really gaining great momentum. Inference is one of the hardest computer science problems. Compiling these gigantic neural network computational graphs into a target device really, really -- has proven to be really, really hard. The models are diverse when you count vision to language to speech. And there are so many different types of models being created that model sizes are doubling every couple of months. The latency expectations are increasing all the time, -- or latency is decreasing all the time. And so, the pressure on inference is really great. The technology pressure is really great. And our leadership there is really pulling ahead. We’re in our seventh generation tensorRT. We, over the course of the last couple of years, developed an inference server. It’s called Triton, has been adopted all over the place. We have several hundred customers now using NVIDIA AI to deploy their AI services. This is in the early innings, and I think this is going to be our largest near-term growth opportunity. So, we’re really firing on all cylinders there between the A100s ramping in the cloud, A100s beginning to ramp in enterprise, and all of our inference initiatives are really doing great.

Jensen Huang:

Yes. John, that’s a good point. I mean, it’s very clear that the inability to go to work is slowing down the adoption of new technology in some of the verticals. Of course, we’re seeing rapid adoption in certain verticals, like for example using AI in health care to rapidly discover new vaccines and early detection with outbreaks and robotic applications. So, warehouses, digital retail, last mile delivery, we’re just seeing just really, really great enthusiasm around adopting new AI and robotics technology. But in some of the more traditional industries, new capabilities and new technologies are slow to deploy. One of the areas that I’m really super excited about is the work that we’re doing in remote work and making it possible for people to collaborate remotely. We have a platform called Omniverse. It’s in early beta. The feedback from the marketplace has been really great. And so, I’ve got a lot more to report to you guys in the upcoming months around Omniverse. And so -- but anyways, I think, when the industry recover, we serve -- our fundamental purpose as a company is to solve the greatest challenges that impact the industry where ordinary computers can’t. And these challenges are -- serve some of the most important applications in the verticals that we address. And they’re not commodity applications. They’re really impactful, needle-moving applications. So, I have every confidence that when the industries recover, things will get designed. Cars will be designed, and planes will be designed, and ships will be designed, and buildings will be designed. And we’re going to see a lot of design, and we’re going to see a lot of simulation. We’re going to see a lot of robotics applications.

C.J. Muse:

You talked about in your prepared remarks limited availability of capacity components. You suggested perhaps a few months to catch up. Curious if you can speak to the visibility that you have for both, gaming and data center into your April quarter?

Colette Kress:

Yes, absolutely.

Operator:

Your next question comes from the line of Stacy Rasgon from Bernstein Research. Your line is open.

Colette Kress:

Sure. Let me try this one, Jensen. Yes, we’ve incorporated that 14th week into our guidance for both, revenue and OpEx. We will likely have incrementally positive impact on revenue, although it is tough to quantify, okay? Our outlook also reflects incremental OpEx for Q4 in primarily two different areas in terms of compensation and depreciation. And given that our employees with such a material power of our OpEx, it will -- it can be close to one-fourteenth of the quarter. Now, when we look a little bit farther, we should think about the incremental positive in both, gaming and data center from that extra week as there hopefully will be extra supply, but not likely as much as one-fourteenth of the quarter of revenue as enterprise demand is essentially project-based, and game demand though is tied to the number of gaming that gamers that might be shopping for the overall holiday. So, again, still very hard for us to determine at this time. Normally, between Q4 and Q1 there is seasonality in gaming, seasonality downward. But, we’ll just have to see as we are still supply-constrained within this Q4 to see what that looks like. From an OpEx standpoint, we’ll probably expect our OpEx to be relatively flattish as we move from Q4 to Q1.

Vivek Arya:

Thanks for taking my question and congratulations on the strong growth. Jensen, my question is on competition from internally designed products by some of your larger cloud customers, Amazon and Google and others. We hear about competition from time to time. And I wanted to get your perspective. Is this a manageable risk? Is the right way to think that they are perhaps using more of your product in their public cloud, but they are moving to internal products for internal workloads? Just how should we think about this risk going forward? Thank you.

Operator:

Your next question comes from the line of Harlan Sur from JP Morgan. Your line is open.

Jensen Huang:

Yes. Let’s see. Our data center business consists of supercomputing centers, which is small, high-performance computing, which is a much larger part of super computing, much larger than supercomputing, and then hyperscale and enterprise, which about 50-50. Of the data center business, the accelerated computing part is not very much associated with digestion than others. It’s much more associated with workloads and our new product cycles, the TCO that we bring in and AI inference, the type of models that the cloud service providers are deploying, whether they’re deploying new AI models, based on deep learning, and how much of it that we -- how much of those workloads that we’ve completed, the porting to our accelerators and ready it for deployment. And so, those are the factors associated with accelerated computing. It’s really about the apps, it’s really about the workloads and really driven by AI. On the other hand, the networking part of our business is more connected to CPU business because they’re much more broad-based. The networking part of our business is driven by this idea of new hyperscale data center architecture called disaggregation. It’s software disaggregation, not necessarily hardware disaggregation. Software disaggregation, where this type of software called Kubernetes orchestrate microservices that are deployed across the data center. So, one service, one application isn’t monolithic running on one computer anymore. It’s distributed across multiple computers and multiple nodes, so that the hyperscale data centers can more easily scale up and scale out according to the workloads and according to the demand on the data center. And so, this disaggregation has caused the networking between the compute nodes to be of all vital importance. And because Mellanox is the lowest-latency, highest-performance, highest-bandwidth network that you can get the TCO benefit at the data center scale is really fantastic. And so, when they’re building out data centers, Mellanox is going to be much more connected to that. In the enterprise side of it, depending on new CPU cycles, it could affect them. If a CPU cycle were to delay a little bit, it would affect them by a quarter, it would pull-in by a quarter -- it would affect them by pull-in of a quarter. And so, those are kind of the dynamics of it. I think, that the net-net of it is that it’s a foregone conclusion at this point that AI is going to be the future of the way software is run. AI is the most powerful technology force of our time. And acceleration is the best path forward. And so, that’s what drives our computing business. And the networking business has everything to do with the way architecture of data centers, cloud data centers, which is architected with micro services now. And that’s what foundationally drives their -- our networking work business demand. And so, we’re really well-positioned in these two fundamental dynamics because as we know, AI is the future and cloud computing is the future. Both of those dynamics are very favorable to us.

Timothy Arcuri:

Thanks a lot. I wanted to ask a question that was asked before in a bit of a different way. If I look at the core business, excluding Mellanox, the core data center business, it was up about 6% sequentially the past two quarters, and your guidance sort of implies up about that much again in January, which is certainly good, and there is some cloud digestion. But, of course you have Ampere still ramping as well, which should be a pretty good tailwind. So, there seems to be some offsetting factors. So, I guess, I wonder if you feel like your core data center revenue is still being constrained right now by some market digestion. And kind of how you sort of balance or handicap these two factors? Thanks.

Operator:

Your next question comes from the line of Aaron Rakers from Wells Fargo. Your line is open.

Jensen Huang:

Long term, every computer in the world will be built like a data center. And every node of a data center is going to be a data center in itself. And the reason for that is because we want the attack surface to be basically zero. And today, most of the data centers are only protected as a periphery. But in the future, if you would like cloud computing to be the architecture for everything and every data center is multi-tenant, every data center is secure, then you’re going to have to secure every single node. And each one of those nodes are going to be a software-defined networking, software-defined storage, and it’s going to have per application security. And so, the processing that it will need to offload, the CPUs, is really quite significant. In fact, we believe that somewhere between 20% to 40% of today’s data centers -- cloud data centers is the capacity, the throughput, the computational load is consumed running basically the infrastructure overhead. And that’s what the DPUs intended -- was designed to do. We’re going to offload that, number one; and number two, we’re going to make every single application secure. And confidential competing, zero trust computing will become a reality. And so, the importance is really quite tremendous. And I believe therefore that every single server in the world will have a DPU inside someday, just because we care so much about security and just because we care so much about throughput and TCO. And it’s really the most cost-effective way of building a data center. And so, I expect our DPU business to be quite large. And so, that’s the reason why we’re putting so much energy into it. It’s a programmable data center on a chip, think of that way, and data center infrastructure on a chip. It is the reason why we’re working with VMware on taking the operating system in the data center, the software-defined operating system in the data center, putting bring it on BlueField. And so, this is a very important initiative for us. I’m very excited about it, as you can imagine.

Ambrish Srivastava:

Yes. Thank you very much. Colette -- and I apologize if I missed it, but for Mellanox, do you expect it to get back to the growth trajectory on a sequential basis in the April quarter? And I’m assuming that the shortfall in the current quarter is from a pull-in from Huawei?

Jensen Huang:

At the highest level, Colette, I think, the -- it’s safe to say that high-speed networking is going to be one of the most important things in cloud data centers as we go forward. And the vast majority of the world’s data center is still built for the traditional hyper-converged architecture, which is all moving over to microservices-based disaggregate --software-defined disaggregated architectures. And that journey is still in its early days. And so, I fully expect future cloud data centers -- all future data centers are going to be connected with high-speed networking inside. They call it east-west traffic. And, all of the traffic will be secured. And so, imagine building firewalls into every single server. And imagine every single transaction, every single transmission inside the data center to be high speed and fully encrypted. And so, pretty amazing amount of computation is going to have to be installed into future data centers. But, that’s an accepted requirement now. And I think, our networking business, Mellanox is in the early innings of growth.

William Stein:

Great. Thanks for taking my question. You’ve given us some pieces of this puzzle. But, I’m hoping maybe you can address directly the sort of SKU-by-SKU rollout of Ampere. We know that we didn’t have a ton of SKUs last quarter. There were more in this quarter that you just announced. Now, you’re doing sort of this refresh it sounds like with double the memory on the A100. Is the T4 going to be refreshed? And if so, when does that happen? And, are there other, either systems or chips that are still waiting for the Ampere refresh that could potentially contribute to an extended cycle as we look at the next year?

Operator:

That concludes our Q&A for today. I now turn the call back to Ms. Jankowski for closing remarks.

Operator:

My apologies.

Operator:

That concludes today’s conference call. You may now disconnect.

Operator:

Good afternoon. My name is David and I will be your conference operator today. At this time, I would like to welcome everyone to NVIDIA’s Financial Results Conference Call. [Operator Instructions] Thank you. Simona Jankowski, you may begin your conference.

Colette Kress:

Thanks, Simona. Q2 was another extraordinary quarter. The world continued to battle the COVID-19 pandemic and most of our employees continue to work from home, but through the team’s agility and dedication, we successfully combined Mellanox into NVIDIA while also delivering a very strong quarter. Revenue was $3.87 billion, up 50% year-on-year, up 26% sequentially and well ahead of our outlook. Starting with gaming, revenue was $1.65 billion, was up 26% year-on-year and up 24% sequentially significantly ahead of our expectations. The upside is broad-based across geographic regions, products and channels. Gaming’s growth amid the pandemic highlights the emergence of a leading form of entertainment worldwide. For example, the number of daily gamers on Steam, a leading PC game online distributor, is up 25% from pre-pandemic levels and NPD reported that U.S. consumer spending on videogames grew 30% in the second calendar quarter to a record $11 billion. NVIDIA’s PCs and laptops are ideal for the millions of people who are now working, learning and gaming at home. At the outset of the pandemic, many retail outlets were closed and demand shifted to online channels. As the quarter progressed and the stores reopened, retail demand picked up, iCafes largely reopened and online sales continued to thrive. Gaming laptop demand is very strong as students and professionals turn to GeForce-based systems to improve how they work, learn, and game from home. We ramped over 100 new models with our OEM partners focused on both premium and mainstream price points. In the premium laptop segment, we delivered unparalleled performance with the GeForce RTX 2080 and the 2070 SUPER GPUs [indiscernible] form factors. We also brought ray tracing to gaming laptops for the first time at price points as low as $999 with the GeForce RTX 2060. In the mainstream segment, we brought the GeForce GTX to laptop price points as low as $699. Momentum continues for our Turing architecture, which enables stunning new visual effects in games and is driving powerful upgrade cycle among gamers. Its RTX technology adds ray tracing and AI to programmable shading and has quickly redefined the new standard for computer graphics. DLSS used the AI capabilities of Turing to boost frame rates by almost 2x, while generating crisp image quality. RTX support in blockbuster games continues to grow, including Mega-Hit DEATH STRANDING, the highly anticipated Cyberpunk 2077 and the upcoming release of Watch Dogs. These games join Minecraft and other major titles that support NVIDIA RTX ray tracing and DLSS. We are in the midst of a 21-day countdown campaign promoting a GeForce special event on September 1, with each day highlighting a year in the history of GeForce. We don’t want to spoil the surprise, but we encourage you to tune in. We are very pleased with the traction of our GeForce NOW cloud gaming service now in its second quarter of commercial availability. GFN offers the richest content to any game streaming service through partnerships with leading digital game stores, including Valve Steam, Epic Games and Ubisoft Uplay. GeForce NOW enables users with underpowered PC, Macs or Android devices to access powerful GPUs to play their libraries of PC games in the cloud, expanding the universe of gamers that we can reach with GeForce. Just yesterday, we announced that GFN is now supported on Chromebooks further expanding our reach into tens of millions of users. In addition to NVIDIA’s own service, GFN is available or coming soon to a number of telecom partners around the world, including SoftBank and KDDI in Japan, Rostelecom and Beeline in Russia, LG U+ in South Korea and Taiwan Mobile. Moving to Pro Vis, in Q2 was $203 million in revenue, down 30% year-on-year and down 34% sequentially, with declines in both mobile and desktop workstations. Sales were hurt by lower enterprise demand amid the closure of many offices around the world. Industries negatively impacted during the quarter, include automotive, architectural engineering and construction, manufacturing, media and entertainment and oil and gas. Initiatives by enterprises to enable remote workers drove demand for virtual and cloud-based graphic solutions. Accordingly, our Q2 vGPU bookings accelerated increasing 60% year-on-year. Despite near-term challenges, we are winning new business in areas such as healthcare, including Siemens, Philips and General Electric and the public sector. We continue to expand our market opportunity with over 50 leading design and creative applications that are NVIDIA RTX-enabled, including the latest release from Foundry, Chaos Group, and Maxon. These applications provide faster ray tracing and accelerated performance, improving creators’ design workflows. The pandemic will have a lasting impact on how we work. Our revenue mix going forward will likely reflect this evolution in enterprise workforce trends with a greater focus on technologies, such as NVIDIA laptops and virtual workstations that enable remote work and virtual collaboration. Moving to automotive, automotive revenue was $111 million, down 47% year-over-year and down 28% sequentially. This was slightly better than our outlook of a 40% sequential decline as the impact of the pandemic was less pronounced than expected, with auto production volumes starting to recover after bottoming in April. Some of the decline is also due to the roll-off of legacy infotainment revenue, which remained a headwind in future quarters. In June, we announced a landmark partnership with Mercedes-Benz, which starting in 2024 will launch software-defined intelligent vehicles across an entire fleet in using end-to-end NVIDIA technology. Mercedes will utilize NVIDIA’s full technology stack, including the DRIVE AGX computer, DRIVE AV autonomous driving software and NVIDIA’s AI infrastructure spanning from the core to the cloud. Centralizing and unifying computing in the car will make it easier to integrate and upgrade advanced software features as they are developed. With over-the-air updates, vehicles can receive the latest autonomous driving and intelligent cockpit features, increasing value and extending majority of ownership with each software upgrade. This is a transformator announcement for the automotive industry making the turning point of traditional vehicles becoming high-performance updatable data centers on wheels. It’s also a transformative announcement for NVIDIA’s evolving business model as the software content of our platforms grows positioning us to build a recurring revenue stream. Moving to data center, data center is a diverse, consist of cloud service providers, public cloud providers, supercomputing centers, enterprises, telecom and industrial edge. Q2 revenue was a record $1.75 billion, up 167% year-on-year and up 54% sequentially. In Q2, we incorporated a full quarter of contribution from the Mellanox acquisition, which closed on April 27, the first day of our quarter. Mellanox contributed approximately 14% of company revenue and just over 30% of data center revenue. Both compute and networking within data center set a record with accelerating year-on-year growth. The biggest news in data center this quarter was the launch of our Ampere architecture. We are very proud of the team’s execution in launching and ramping this technological marvel especially amid the pandemic. The A100 is the largest chip ever made with 54 billion transistors. It runs our full software stack for accelerating the most compute-intensive workloads. Our software leases include CUDA 11, the new versions of over 50 CUDA-X libraries and a new application framework for major AI workloads, such as Jarvis for conversational AI and Merlin for deep recommender systems. The A100 delivers NVIDIA’s greatest generational leap ever, boosting AI performance by 20x over its predecessor. It is also our first universal accelerator unifying AI training and inference and powering workloads, such as data analytics, scientific computing, genomics, edge video analytics, 5G services and graphics. The first Ampere GPU, A100, has been widely adopted by all major server vendors and cloud service providers. Google Cloud platform was the first cloud customer to bring it to market making it the fastest GPU to come to the cloud in our history. And just this morning, Microsoft Azure announced the availability of massively scalable AI clusters, which are based on the A100 and interconnected with 200-gigabyte-per-second Mellanox InfiniBand networking. A100 is also getting incorporated into offerings from AWS, Alibaba Cloud, Baidu Cloud and Tencent Cloud. And we announced that the A100 is going to market with more than 50 servers from leading vendors around the world, including Cisco, Dell, Hewlett-Packard Enterprise and Lenovo. Adoption of the A100 into leading server makers offerings is faster than any prior launch, with 30 systems expected this summer and over 40 more by the end of the year. The A100 is already winning industry recognition in the latest A100 training benchmark, MLPerf 0.7 NVIDIA set 16 records, sweeping all categories for commercially available solutions in both per chip and outscale performance based on the A100. MLPerf offers the industry’s first and only objective AI benchmark. Since the benchmark was introduced 2 years ago, NVIDIA has consistently delivered leading results and record performance for both training and inference. NVIDIA also topped the chart in the latest TOP500 list of the fastest supercomputers. The ranking, released in June, showed that 8 of the world’s top 10 supercomputers use NVIDIA GPUs, NVIDIA’s networking or both. They include the most powerful systems in the U.S. and Europe. NVIDIA now combined with Mellanox powers two-thirds of the top 500 systems on the list compared with just less than a half for the two companies in total 2 years ago. In energy efficiencies, systems using NVIDIA GPUs are pulling away from the pack. On average, they are nearly 2.8x more powerful and efficient than systems without NVIDIA GPUs measured by gigaflops per watt. The incredible performance and efficiency of the A100 GPU is best amplified by NVIDIA’s own new Selene supercomputer, which debuted as number seven on the TOP500 list and is the only top 100 systems to cross the 20 gigaflops per watt barrier. Our engineers were able to assemble Selene in less than 4 weeks using NVIDIA’s open modular DGX SuperPOD reference architecture instead of the typical build time of months or even years. This is the system that we will use to win the MLPerf benchmarks and it is a reference design that’s available for our customers to quickly build a world class supercomputer. We also brought GPU acceleration to data analytics, one of the largest and fastest growing enterprise workloads. We enabled end-to-end acceleration of the entire data analytics workload pipeline for the first time with NVIDIA’s GPUs and software stack in the latest version of Apache Spark released in June. Spark is the world’s leading data analytics platform used by more than 500,000 data scientists and 16,000 enterprises worldwide. And we have two major milestones to share. We have now shipped a cumulative total of 1 billion CUDA GPUs and the total number of developers in the NVIDIA ecosystem just reached 2 million. It took over a decade to reach the first million and less than 2 years to reach the second million. Mellanox has fantastic results across the Board in its first quarter as part of NVIDIA. Mellanox revenue growth accelerated with strength across Ethernet and InfiniBand products. Our Ethernet shipments reached a new record. Major hyperscale build drove the upside in the quarter as growth in cloud computing and AI is fueling increased demand for high-performance networking. Mellanox networking was a critical part of several of our major new product introductions this quarter. These include the DGX AI system, the DGX SuperPOD clusters for our Selene supercomputer and the EGX Edge AI platform. We also launched the Mellanox ConnectX-6 Ethernet NIC, the 11th generation product of the ConnectX family and is designed to meet the needs of modern cloud and hyperscale data centers, where 25, 50 and 100 gigabyte per second is becoming the standard. We expanded our switch networking capabilities with the addition of Cumulus Networks, a privately held network software company that we purchased in June. Cumulus augments our Mellanox acquisition in building out open modern data center. The combination of NVIDIA accelerated computing, Mellanox networking and Cumulus software enables data centers that are accelerated, disaggregated and software-defined to meet the exponential growth in AI, cloud and high-performance computing. Moving to the rest of the P&L. Q2 GAAP gross margin was 58.8% and non-GAAP gross margin was 66%. GAAP gross margin declined year-on-year and sequentially due to costs associated with the Mellanox acquisition. Non-GAAP gross margins increased by almost 6 points year-on-year, reflecting a shift in product mix with higher data center sales and lower automotive sales. Q2 GAAP operating expenses were $1.62 billion and non-GAAP operating expenses were $1.04 billion, up 67% and 38% from a year ago, respectively. Q2 GAAP EPS was $0.99, up 10% from a year earlier. Non-GAAP EPS was $2.18, up 76% from the year ago. Q2 cash flow from operations was $1.57 billion. With that, let me turn to the outlook for the third quarter of fiscal 2021. We expect revenue to be $4.4 billion, plus or minus 2%. With that, we expect gaming to be up just over 25% sequentially, with data center to be up in the low to mid-single digits sequentially. We expect both pro vis and auto to be at similar levels out in Q2. GAAP and non-GAAP gross margins are expected to be 62.5% and 65.5% respectively plus or minus 50 basis points. GAAP and non-GAAP operating expenses are expected to be approximately $1.54 billion and $1.09 billion respectively. Full year GAAP and non-GAAP OpEx is tracking in line with our outlook of $5.7 billion and $4.1 billion respectively. GAAP and non-GAAP OI&E are both expected to be expense of approximately $55 million. GAAP and non-GAAP tax rates are both expected to be 8%, plus or minus 1%, excluding discrete items. Capital expenditures are expected to be approximately $225 million to $250 million. Further financial details are included in the CFO commentary and other information available on our IR website. In closing, let me highlight upcoming events for the financial community. We will be at the BMO Virtual Technology Summit on August 25, Citi’s 2020 Global Technology Conference on September 9, Deutsche Bank’s Technology Conference on September 14 and the Evercore’s Virtual Memo Forum, The Future of Mobility, on September 21. We will also host a financial analyst Q&A with Jensen on October 5 as part of our next virtual GTC. Our earnings call to discuss our third quarter’s results is scheduled for Wednesday, November 18. We will now open the call for questions. Operator, would you please poll for questions? Thank you.

Vivek Arya:

Thanks for taking my question and congratulations on the strong growth and execution. Jensen, I am wondering how much of the strength that you are seeing in gaming and data center is maybe more temporary because of COVID or some customer pull-ins in the data center or so forth? And how much of it is more structural and more secular that can continue even once we get into hopefully sooner rather than later into a more normalized period for the industry?

Operator:

Your next question comes from the line of Timothy Arcuri with UBS. Your line is open.

Jensen Huang:

Yes. We push architecture really hard. And the way we push architecture is we start from the system. But we believe that the future computer company is a data center-scale company. The computing unit is no longer a microprocessor or even a server or even a cluster. The computing unit is an entire data center now. And as I was explaining it to Vivek just now that a microservice that we are enjoying hundreds of billions of transactions a day, those are broken up into a whole bunch of microservices that are running across the entire data center. And so, the data center is running – the entire data center is running an application. I mean, that’s pretty remarkable thing. And that happened in the last several years. We were ahead of this trend, and we recognized that, as a computing company, we have to be a data center scale company and we architect from that starting. If you look at our architecture, we were the first in the world to create the concept of an NVLink, with eight processors being fully synchronized across the computing node, and we created the DGX. We recognize the importance of high-speed networking and low-latency networking, and that’s why we bought Mellanox. And the amount of software that we invented along the way to make it possible for low-latency communications, whether it’s GPUDirect or, recently, the invention of GPUDirect Storage, all of that technology was inspired by the idea that you have to think about the data center all in one holistic way. And then in the last – in this current generation with Ampere, we invented the world’s first multi-instance GPU. We invented the world’s first multi-instance GPU, which means that our Ampere GPU could simultaneously be one GPU or, with NVlink, 8 GPUs combined, working together. So you could think of them as being titled, so those 8 GPUs are working harmoniously together or each one of the GPUs could fractionalize itself, if you don’t need that much GPU working on your workload, fractionalize into a multi-GPU instance, we call the MIG, a little tiny instance. And each one of those tiny instances are more powerful and more performant than our entire Volta GPU lap time. And so, whether you like to fractionalize the GPU, which happens oftentimes, create a larger GPU using NVLink, or create an even larger GPU, the size of DGX POD, connected together with high-speed, low-latency networking with Mellanox, we could architect it any way you would like. You made a comment about – you asked a question about ARM. We have been a long-term partner of ARM, and we use ARM in a whole bunch of applications. And whether it’s autonomous driving or a robotics application, the Nintendo Switch, console business that we are in. And then, recently, we brought CUDA to ARM and to bring accelerated computing to ARM. And so, we worked with the ARM team very closely. They are really great guys. And one of the specials about the ARM architecture that you know very well is that it’s incredibly energy-efficient. And because it’s energy-efficient, it has the headroom to scale into very high-performance levels over time. And so, anyways, we love working with the ARM guys.

Aaron Rakers:

Yes. Thanks for taking the question. Congratulations on the quarter. Just building on some prior questions. The first one, I was just curious if you could help us appreciate kind of the installed base of the gaming GPU business, relative to where we are at the Turing upgrade cycle, what do we see still on prior generations, be it Pascal or before? And then secondly, I was wondering, Colette, could you just give me a kind of updated commentary or views on visibility in the data center business? How that – how has that changed over the last 3 months? What does that look like as you look through the back half of the calendar year? Thank you.

Aaron Rakers:

And then on the data center visibility?

Operator:

Your next question comes from the line of C.J. Muse with Evercore ISI. Your line is open.

Jensen Huang:

Yes. Thanks so much, C.J. We are expecting a really strong second half for gaming. I think this may very well be one of the best gaming seasons ever. And the reason for that is because PC gaming has become such a large format. The combination of amazing games like Fortnite and Minecraft and because of the way people game now, their gaming and their e-sporting, even F1 is an e-sport now, they are hanging out with friends. They are using it to create other content. They are using, game captures, create art. They are sharing it with the community. It’s a broadcast medium. The number of different ways you could game has just really, really exploded. And it works on PCs because all the things that I described, require cameras or keyboards or streaming systems and – but it requires an open system that is multitasking. And so, the PC has just become such a large platform for gaming. And the second thing is that RTX, it’s a home run. We really raised the bar with computer graphics, and the games are so beautiful, and it’s really, really the next level. It’s not been this amazing since we introduced programmable shaders about 15 years ago. And so, for the last 15 years, we have been making programmable shaders better and better and better, and it has been getting better. But there’s never been a giant leap like this. And RTX brought both artificial intelligence as well as ray tracing to PC gaming. And then the third factor is the console launch. There’s – people are really – the game developers are really gearing up for a big leap. And because of the gaming – because how vibrant the gaming market is right now and how many people around the world is depending on gaming at home, I think it’s going to be the most amazing season ever. We are already seeing amazing numbers from our console partner, Nintendo. Switch has – about to sell more than Super Nintendo, more than all the Famicom. I mean – which was one of the best gaming consoles of all time. I mean, they are on their way to make Switch the most successful gaming platform of all time. And so, I am super excited for them. And so, I think it’s going to be quite a huge second half of the year.

Jensen Huang:

Colette, I felt like I didn’t – I missed C.J.’s second question. Can we jump on and answer it?

Jensen Huang:

Yes, that’s the one. Yes.

C.J. Muse:

Okay.

Toshiya Hari:

Hi. Good afternoon and thank you so much for taking the question. I had one for Jensen and another one for Colette. Jensen, just following-up on the data center business. As you probably know, quite a few of your peers have been talking about potential digestion of capacity on the part of your hyperscale customers over the next call it, 6 to 12 months. Curious, is that something that you think about, worry about in your data center business? Or do you have enough idiosyncratic growth drivers like the A100 ramp? And I guess the breadth that you have built within your data center business across compute and networking, are those enough for you to buck the trend within data center over the next 6 to 12 months? And then the second one for Colette, just on gross margins, you are guiding October quarter gross margins down 50 basis points sequentially. Based on the color that you provided for the individual segments, it looks like mix remains pretty positive. So just curious what’s driving the marginal decline in gross margins in the October quarter? Thank you.

Colette Kress:

Okay. And, Toshiya, you asked a question regarding our guidance going forward regarding gross margin. And within our Q3 guidance, we have just a small decline in our gross margin from Q2. Most of that is really associated with mix, but also a little bit in terms of the ramping of our new Ampere architecture products that we have. So, keep in mind, our data center will likely be a lower percentage of total revenue, given the strong overall gaming growth that we expect between Q2 and Q3. Within that gaming growth, keep in mind, consoles are also included, which will continue to be below our company totals average gross margin, and that is expected to be up strongly quarter over quarter for our overall console shipments. We are going to be ramping those new architectures over time when we have the ability to expand our gross margin as Ampere GPUs mature, too.

Stacy Rasgon:

Hi, guys. Thanks for taking my question. I wanted to dig into data center a little bit. This is a question for Colette. So, in the quarter, ex Mellanox, data center was up, core data center, maybe 6%, 7%. The guide looks to be roughly similar to that into Q3. Can you talk to us a little bit about what’s driving the trajectory? Are you more demand- or more supply limited at this point? What does your supply situation look like? And what are the lead times especially on the A100 products for data center look like at this point? Like if you have more capacity available, do you think you’d have like a stronger trajectory than you have right now?

Jensen Huang:

Well, when we are ramping, we sure love to have more and sooner. And – but this is our plan, and we are executing to the plan. It is a very complicated product, as Colette mentioned. It is the most complicated.

Colette Kress:

Yes. So, in terms of moving from Q2 to Q3, we believe that most of the actual growth that we will receive in that single – low single-digits to mid-single-digit growth will likely stem from NVIDIA compute will be the largest driver of that.

Joseph Moore:

Great. Thank you. I wonder if I could ask a longer-term question about the – how you guys see the importance of process technology. There’s been a lot of discussion around that in the CPU domain. But you guys haven’t really felt the need to be first on seven-nanometer, and you have done very well. Just how important do you think it is to be early in the new process node? And how does that factor into the cycle of innovation at NVIDIA?

Operator:

We have no further questions at this time. Jensen Huang, I turn the call back over to you.

Operator:

This concludes today’s conference call. You may now disconnect.

Operator:

Good afternoon. My name is Josh, and I will be your conference operator today. At this time, I would like to welcome everyone to NVIDIA's Financial Results Conference call. [Operator Instructions]. Simona Jankowski, you may begin your conference.

Jensen Huang:

Thanks, Simona. Before Colette describes our quarterly results, I'd like to thank those who are on the front lines of this crisis, first responders, health care workers, service providers, who inspires every day with their bravery and selflessness. I also want to acknowledge the incredible efforts of our colleagues here at NVIDIA. Despite many challenges, they have barely broken stride during one of the busiest periods in our history. Our efforts related to the virus are focused in 3 areas. First, we're taking care of our families and communities. We've pooled in raises by 6 months to put more money in our employees' hands, and NVIDIA and our people have donated thus far more than $10 million to those in need. Second, we're using NVIDIA's unique capabilities to fight the virus. A great deal of science being done on COVID-19 uses NVIDIA technology for acceleration when every second counts. Some of the many examples, including sequencing the virus, analyzing drug candidates, imaging the virus at molecular resolution with cryo-electron microscopy and identifying elevated body temperature with AI cameras. And third, because COVID-19 won't be the last killer virus, we need to be ready for the next outbreak. NVIDIA technology is essential for the scientific community to develop an end-to-end computational defense system, a system that can detect early, accelerate the development of a vaccine, contain the spread of disease and continuously test and monitor. We are racing to deploy the NVIDIA Clara computational health care platforms, Clara Parabricks can accelerate genomics analysis from days to minutes. Clara Imaging will continue to partner with leading research institutes to develop state-of-the-art AI models to detect infections, and Clara Guardian will connect AI to cameras and microphones and hospitals to help overloaded staff watch over patients. We completed the acquisition of Mellanox on April 27. Mellanox is now NVIDIA's networking brand and business unit and will be reported as part of our data center market platform, and Israel is now one of NVIDIA's major technology centers. The new NVIDIA has a much larger footprint in data center computing, end-to-end and full-stack expertise in data center architectures and tremendous scale to accelerate innovation. NVIDIA Mellanox are a perfect combination and position us for the major forces shaping the IT industry today, data center scale computing and AI. From micro service cloud applications to machine learning and AI, accelerated computing and high-performance networking are critical to modern data centers. Previously, a CPU compute node was the unit of computing. Going forward, the new unit of computing is an entire data center. The basic computing elements are now storage servers, CPU servers and GPU servers, and are composed and orchestrated by hyperscale applications that are serving millions of users simultaneously. Connecting these computing elements together is the high-performance Mellanox networking. This is the era of data center scale computing. And together, NVIDIA Mellanox can architect end to end. Mellanox is an extraordinary company, and I'm thrilled that we're now one force to invent the future together. Now let me turn the call over to Colette.

Operator:

[Operator Instructions]. Aaron Rakers with Wells Fargo.

Colette Kress:

Thanks, Will, for the question. You are correct. We have indicated a couple of quarters ago that we were starting to see improved visibility after we came out of the digestion period in the prior overall fiscal year. As we move into Q2, we still have visibility and solid visibility into our Q2 results for overall data centers. So at this time, I'd say they are relatively about the same of what we had seen going into the Q1 period. And we think that is a true indication of their excitement about our platform and most particularly our excitement regarding A100, and that's launched and its additional products. Now regarding the second half of the year, as you know, we have seen broad-based growth in both the hyperscale and the vertical industries, both of them in terms of at record levels. In our Q1 results. And we see in terms of inferencing continuing to grow as well, as well as we're also expanding in terms of edge AI. Our strong demand of the A100 products, including the Delta Board, but also in terms of our DGXs, was just starting an initial ramp. However, we do guide only 1 quarter at a time. So it's still a little bit too early for us to give a true certainty in terms of the macro situation that's in front of us. But again, we feel very good about the demand for A100.

Stacy Rasgon:

I first wanted to follow-up on your gaming commentary. You sort of mentioned a couple of offsets. COVID potentially still a headwind, e-tail or tailwind and maybe offsetting each other. Were you trying to suggest that those did offset completely and gaming was kind of flattish into Q2? Because I know it has a typical seasonal pattern switches typically up. I guess what were you trying to say with those kind of factors? And what are the kinds of things we should be thinking about when it comes to seasonality, Colette, into Q2 around that business segment?

Stacy Rasgon:

Yes. That's right.

Jensen Huang:

No, that was great. That was fantastic.

Colette Kress:

Yes. I think when we think about that sequential growth, we'll probably be in the low -- moving up to probably the mid-single digits in terms of -- that's what our guidance right now, and we'll just have to see how the quarter goes.

Jensen Huang:

Stacy, the thing that I would add is this. I would say, I think the guidance is exactly what Colette mentioned. But if you look at the big picture, there's a few dynamics that are working really well in our favor. First, of course, is that RTX and ray tracing is just the home run. Minecraft was phenomenal. We have 33 games in the pipe that has already been announced or shipping. Just about every game developers signed on to RTX and ray tracing, and I think it's a foregone conclusion that this is the next generation. This is the way computer graphics is going to be in the future. And so I think RTX is a home run. The second, the notebooks that we create is just doing great. We got 100 notebooks in gaming. We have 75 notebooks designed for either mobile workstations or what we call NVIDIA studio for designers and creators. And the timing was just perfect. With everybody needing to stay at home, the ability to have a mobile gaming platform and a mobile workstation, it was just perfect timing. And then, of course, you guys know quite well that our Nintendo Switch is doing fantastic. There are the Top 3 games in the world. The top games in the world today are Fortnite, Minecraft and Animal Crossing. All 3 games are NVIDIA platforms. And so I think we have all the dynamics working in our favor. And then we just got to see how it turns out.

Joseph Moore:

I wanted to ask about the rollout of Ampere how quickly does that roll in the various segments between hyperscale as well as on the DGX side as well as on the HPC side. And is it a smooth transition? Is there -- I remember when you launched Volta, there was a little bit of a transitional pause. Just can you tell us how you see that ramping up with the different customer segments?

Operator:

Your next question comes from Vivek Arya with Bank of America.