AI infrastructure demand remains supply-constrained, with Blackwell ramping from Q2 and potentially extending growth into 2025.
| Quarter | Q1 FY25 |
|---|---|
| Results reported | 2024-05-22 |
| Length | 65 minutes |
| Speakers | Simona Jankowski, Head of Investor Relations; Colette Kress, Executive Vice President and CFO; Jensen Huang, President and CEO |
Results, guidance and Q&A analysis for this call
Regina (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 first 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. If you would like to ask a question during this time, simply press * followed by the number one on your telephone keypad. If you would like to withdraw your question, press *1. Again, thank you. Simona Jankowski, you may begin your conference.
Simona Jankowski (Head of Investor Relations)
Thank you. Good afternoon, everyone, and welcome to Nvidia's conference call for the first quarter of fiscal 2025. With me today from NVIDIA are Jensen Huang, President and Chief Executive Officer, and Colette Crest, Executive Vice President and Chief Financial Officer. I'd like to remind you that our call is being webcast live on Nvidia's Investor Relations website. The webcast will be available for replay until the conference call to discuss our financial results for the second quarter of fiscal 2025. The content of today's call is Nvidia's property. It can be reproduced or transcribed without a prior written consent. During this call, we may make forward-looking statements based on current expectations. These are subject to a number of significant risks and uncertainties and our actual results may differ materially. For a discussion of factors that could affect our future financial results and business, please refer to the disclosure in today's earnings release, our most recent Forms 10K and 10Q and the reports that we may file on Form 8K with the Securities and Exchange Commission. All our statements are made as of today, May 22nd, 2024. Based on information currently available to us. Except as required by law, we assume no obligation to update any such statements. During this call, we will discuss non GAAP financial measures. You can find a reconciliation of these non GAAP financial measures to GAAP financial measures in our CFO commentary which is posted on our website. Let me highlight some upcoming events. On Sunday, June 2nd, ahead of the Computex technology trade show in Taiwan, Jensen will deliver a keynote which will be held in person in Taipei as well as streamed live. And on June 5th, we will present at the Bank of America Technology Conference in San Francisco. With that, let me turn the call over to Colette.
Colette Kress (Executive Vice President and CFO)
Thanks, Simona. Q1 was another record quarter. Revenue of 26 billion was up 18% sequentially and up 262% year on year and well above our outlook of 24 billion. Starting with data center. Data center revenue of 22.6 billion was a record, up 23% sequentially and up 427% year on year, driven by continued strong demand for the NVIDIA Hopper GPU computing platform. Compute revenue grew more than 5X and networking revenue more than 3X from last year. Strong sequential data center growth was driven by all customer types led by enterprise and consumer Internet companies. Large cloud providers continue to drive strong growth as they deploy and ramp NVIDIA AI infrastructure at scale and represented the mid 40s as a percentage of our data center revenue. Training and inferencing AI on NVIDIA CUDA is driving meaningful acceleration in cloud rental revenue growth, delivering an immediate and strong return on cloud providers investment. For every $1.00 spent on NVIDIA AI infrastructure, cloud providers have an opportunity to earn $5 in GPU instant hosting revenue over 4 years. NVIDIA Switch software stack and ecosystem and tight integration with cloud providers makes it easy for end customers up and running on NVIDIA GPU instances in the public cloud. For cloud rental customers, NVIDIA GPUs offer the best time to train models, the lowest cost to train models, and the lowest cost to inference large language models. For public cloud providers, NVIDIA brings customers to their cloud, driving revenue growth and returns on their infrastructure investments. Leading LLM companies such as Open AI, Adapt, Anthropic, Character AI, Cohere Data Bricks, DeepMind, Meta, Mistral, XAI and many others are building on NVIDIA AI in the cloud. Enterprises drove strong sequential growth in data center this quarter. We supported Tesla's expansion of their training AI cluster to 35,000 H 100 GPUs. Their use of NVIDIA AI infrastructure paved the way for the breakthrough performance of FSD version 12, their latest autonomous driving software based on vision video Transformers while consuming significantly more computing or enabling dramatically better autonomous driving capabilities and propelling significant growth for NVIDIA AI infrastructure across the automotive industry. We expect automotive to be our largest enterprise vertical within data center this year, driving a multi billion revenue opportunity across on Prem and cloud consumption. Consumer Internet companies are also a strong growth vertical. A big highlight this quarter was Meta's announcement of LAMA Three, their latest large language model, which was trained on the cost of 24,000 H 100 GPUs. LAMA 3 powers Meta AIA new AI assistant Available on Facebook, Instagram, WhatsApp and Messenger Lama 3 is openly available and has kick started a wave of AI development across industries. As generative AI makes its way into more consumer Internet applications, we expect to see continued growth opportunities as inference scales both with model complexity as well as with the number of users and number of queries per user, driving much more demand for AI compute. In our trailing 4 quarters, we estimate that inference drove about 40% of our data center revenue. Both training and inference are growing significantly. Large clusters like the ones built by Meta and Tesla are examples of the essential infrastructure for AI production, what we refer to as AI factories. These next generation data centers host advanced full stack accelerated computing platforms where the data comes in and intelligence comes out. In Q1, we worked with over 100 customers building AI factories ranging in size size from hundreds to 10s of thousands of GPUs, with some reaching 100,000 GPUs. From a geographic perspective, data center revenue continues to diversify as countries around the world invest in sovereign AI. Sovereign AI refers to a nation's capabilities to produce artificial intelligence using its own infrastructure, data, workforce and business networks. Nations are building up domestic computing capacity through various models. Some are procuring and operating sovereign AI clouds in collaboration with state owned telecommunication providers or utilities. Others are sponsoring local cloud partners to provide a shared AI computing platform for public and private sector use. For example, Japan plans to invest more than 740 million in key digital infrastructure providers, including KDDI, Sakura Internet and SoftBank to build out the nation's sovereign AI infrastructure. France based Scale Way, a subsidiary of the Iliad Group, is building Europe's most powerful cloud native AI supercomputer. In Italy, Swisscom Group will build the nation's first and most powerful NVIDIA DTX powered supercomputer to develop the first LLM natively trained in the Italian language. And in Singapore, the National Supercomputer Center is getting upgraded with NVIDIA Opera GPUs. Why? Singtel is building Nvidia's accelerated AI factories across Southeast Asia. Nvidia's ability to offer end to end compute to networking technologies, full stack software, AI expertise and rich ecosystem of partners and customers allows sovereign AI and regional cloud providers to jump start their country's AI ambitions from nothing the previous year. We believe sovereign AI revenue can approach the high single digit billions this year. The importance of AI has caught the attention of every nation. We ramped new products designed specifically for China that don't require an import control license. Our data center revenue in China is down significantly from the level prior to the imposition of the new export control restrictions in October. We expect the market in China to remain very competitive going forward. From a product perspective, the vast majority of compute revenue was driven by our Hopper GPU architecture. Demand for Hopper during the quarter continues to increase. Thanks to CUDA algorithm innovation, we've been able to accelerate LLM inference on H100 by up to 3X, which can translate to a 3X cost reduction for serving popular models like Llama 3. We started sampling the H200 in A in Q1 and are currently in production with shipments on track for Q2. The first H 200 system was delivered by Jensen to Sam Altman and the team at Open AI and powered their amazing GPT 4 O demos last week. H200 nearly double S the inference performance of H100, delivering significant value for production deployments. For example, using Bama 3, which with 700 billion parameters, a single NVIDIA HGX H200 server can deliver 24,000 tokens per second, supporting more than 2400 users at the same time. That means for every $1.00 spent on NVIDIA HTX H200 servers at current prices per token, an API provider serving Llama 3 tokens can generate $7.00 in revenue. Over 4 years. With ongoing software optimizations, we continue to improve the performance of NVIDIA AI infrastructure for serving AI models. While supply for H100 continue to improve, we are still constrained on H200. At the same time, Blackwell is in full production. We are working to bring up our system and cloud partners for global availability later this year. Demand for H200 and Blackwell is well ahead of supply and we expect demand may exceed supply well into next year. Grace Hopper Superchip is shipping and volume. Last week at the International Supercomputing Conference, we announced that 9 new supercomputers worldwide are using Grace Hopper for combined 200 exoplops of energy efficient AI processing power delivered this year. These include the Alps supercomputer at the Swiss National Supercomputing Center, the fastest AI supercomputer in Europe, Zombard AI at the University of Bristol in the UK, and Jupiter in the ULUC supercomputing Center in Germany. We are seeing an 80% attach rate of GRACE to Hopper in supercomputing to it to its high energy efficiency and performance. We are also proud to see supercomputers powered with Grace Hopper take the number one, the number 2 and the number three spots of the most energy efficient supercomputers in the world. Strong networking year on year growth was driven by incentive bad. We experienced a modest sequential decline, which was largely due to the timing of supply with demand well ahead of what we were able to ship. We expect networking to return to sequential growth in Q2. In the first quarter, we started shipping our new Spectrum X Ethernet networking solution optimized for AI from the ground up. It includes our Spectrum 4 switch, Bluefield 3 DPU and new software technologies to overcome the challenges of AI on Ethernet to deliver 1.6 X higher networking performance for AI processing compared with traditional Ethernet. Spectrum X is ramping and volume with multiple customers including a massive 100,000 GPU cluster. Spectrum X opens a brand new market to NVIDIA networking and enables Ethernet only data centers to accommodate large scale AI. We expect Spectrum X to jump to a multi billion dollar product line within a year. At GTC in March, we launched our next generation AI factory platform Blackwell. The Blackwell GPU architecture delivers up to 4X faster training and 30X faster inference than the H100 and enabled real time generative AI on trillion parameter large language models back Blackwell is a giant leap with up to 25 X lower TCO and energy consumption than Hopper. The Blackwell platform includes the 5th generation and be linked with a multi GPU spine and new InfiniBand and Ethernet switches. The X800 series designed for a trillion parameter scale AI, Blackwell is designed to support data centers universally, from hyperscale to enterprise training to inference, X86 to Gray CPU, Ethernet to InfiniBand networking, and air cooling to liquid cooling. Blackwell will be available in over 100 OEM and ODM systems at large, more than double the number of hoppers launch and representing every major computer maker in the world. This will support fast and broad adoption across the customer types, workloads and data center environments. In the first year shipments Blackwell time to market customers include Amazon, Google, Meta, Microsoft, Opening Eye, Oracle, Tesla and XAI. We announced a new software product with the introduction of NVIDIA and Friends Micro Services or NIM. NEMS provides secure and performance optimized containers powered by NVIDIA CUDA acceleration in network computing and inference software including Triton and Print server and Tensor RTLLM with industry standard AP is for broad range of use cases including large language models for text, speech, imaging, vision, robotics, genomics and digital biology. They envelope enable developers to quickly build and deploy generative AI applications using leading models from NVIDIA AI 21, Adapt, Cohere, Getty Images and Shutterstock and open models from Google, Hugging Face, Meta, Microsoft, Mistral, AI, Snowflake and Stability AI. Mims will be offered as part of our NVIDIA AI enterprise software platform for production deployment in the cloud or on Prem. Moving to gaming and Aipcs. Gaming revenue of 2.65 billion was down 8% sequentially and up 18% year on year, consistent with our outlook for a seasonal decline. The GE Force RTX Supers GPUs market reception is strong and end demand and channel inventory remain healthy across the product range. From the very start of our AI journey, we equipped G Force RTXGPUS with CUDA Tensor cores. Now, with over 100 million of an installed base, G Force RTXGPUS are perfect for gamers, creators, AI enthusiasts and offer unmatched performance for running generative AI applications on P CS. NVIDIA has full technology stack for deploying and running fast and efficient generative AI inference on G Force RTXP CS sensor. RT LLM now accelerates Microsoft 53 Mini model and Google's Gemma 2B and 7B models, as well as popular AI frameworks including Lang Chang and Llama Index. Yesterday NVIDIA and Microsoft announced AI performance optimizations for Windows to help run LLMS up to 3X faster on NVIDIA GeForce RTXAIPCS and top game developers including Nettie's games, Tencent and Ubisoft for embracing NVIDIA Avatar Character Engine to create lifelike avatars to transform interactions between gamers and non playable characters. Moving to Provis, revenue of 427,000,000 was down 8% sequentially and up 45% year on year. We believe generative AI and Omniverse industrial digitalization will drive the next wave of professional visualization growth. At GTC, we announced new Omniverse Cloud AP is to enable developers to integrate Omniverse industrial digital twin and simulation technologies into their applications. Some of the world's largest industrial software makers are adopting these. AP is including ANSYS, Cadence 3D, Excite at Dassault Systems brand and Siemens and developers can use them to stream industrial digital twins with spatial computing devices such as Apple Vision Pro. Omniverse cloud APIs will be available on Microsoft Azure later this year. Companies are using Omniverse to digitalize the workflows. Omniverse powered digital twins enable Wistron, one of our manufacturing partners to reduce end to end production cycle times by 50% and defect rates by 40%. And BYD, the world's largest electric vehicle maker is adopting Omniverse for virtual factory planning and retail configurations. Moving to automotive, revenue was 329 million, up 17% sequentially and up 11% year on year. Sequential growth was driven by the ramp of AI cockpit solutions with global OEM customers and strength in our self driving platforms. Year on year growth was driven primarily by self driving. We supported Xiaomi in the successful launch of its first electric vehicle, the SU-7 sedan built on the NVIDIA Dr. Oren, our AI car computer for software defined AV slates. We also announced a number of new design wins on NVIDIA Drive for the successor to Oren powered by the new NVIDIA Blackwell architecture with several leading EV makers including BYDX, P/E, ACS, Ion, Hyper and Neuro. Dr. Thorough is slated for production vehicles starting next year. OK, moving to the rest of the PNL gap gross margin expanded sequentially to 78.4% and non gap gross margins to 78.9% on lower inventory charges. As noted last quarter, both Q4 and Q1 benefited from favorable component costs. Sequentially, GAAP operating expenses were up 10% and non GAAP operating expenses were up 13%, primarily reflecting higher compensation related costs and increase compute and infrastructure investments. In Q1, we returned 7.8 billion to shareholders in the form of share repurchases and cash dividends. Today we announce A10 for one split of our shares with June 10th as the first day of trading. On a split adjusted basis, we are also increasing our dividend by 150%. Let me turn to the outlook for the second quarter. Total revenue is expected to be 28 billion ± 2%. We expect sequential growth in all market platforms. Gap and non GAAP gross margins are expected to be 74.8% and 75.5% respectively ±50 basis points consistent with our discussion last quarter. For the full year, we expect gross margins to be in the mid 70s percent range. Gap and non GAAP operating expenses are expected to be approximately 4 billion and 2.8 billion respectively. Full year OpEx is expected to grow in the low 40% range. Gap and non GAAP other income and expenses are expected to be an income of approximately 700, excuse me, of approximately 300 million excluding gains and losses from non affiliated investments. GAAP and non GAAP tax rates are expected to be 17% ± 1% excluding any discrete items. Further financial details are included in the CFO commentary and other information available in our IR website. I would like to now turn it over to Jensen as he would like to make a comments.
Jensen Huang (President and CEO)
Thanks Colette. The industry is going through a major change. Before we start Q&A, let me give you some perspective on the importance of the transformation. The next industrial revolution has begun. Companies and countries are partnering with NVIDIA to shift the trillion dollar installed base of traditional data centers to accelerated computing and build a new type of data center AI factories to produce a new commodity. Artificial intelligence. AI will bring significant productivity gains to nearly every industry and help companies be more cost and energy efficient while expanding revenue opportunities. CSPS were the first generative AI movers with NVIDIA CS PS accelerated workloads to save money and power the tokens generated by NVIDIA Hopper. Dr. revenues for their AI services and NVIDIA Cloud instances attract rental customers from our rich ecosystem of developers. Strong and accelerated demand, accelerating demand for generative AI training and inference on the Harper platform propels our data center group. Training continues to scale as models learn to be multimodal, understanding text, speech, images, video and 3D, and learn to reason and plan. Our inference workloads are growing incredibly with generative AI. Inference, which is now about fast token generation at massive scale has become incredibly complex. Generative AI is driving a from foundation up full stack computing platform shift that will transform every computer interaction. From today's information retrieval model, we are shifting to an answers and skills generation model of computing. AI will understand context and our intentions, be knowledgeable, reason, plan and perform tasks. We are fundamentally changing how computing works and what computers can do. From general purpose CPU to GPU accelerated computing, from instruction driven software to intention understanding models, from retrieving information to performing skills, and at the industrial level from producing software to generating tokens. Manufacturing digital intelligence token generation will drive a multi year build out of AI factories. Beyond cloud service providers, generative AI has expanded to consumer Internet companies and enterprise sovereign AI, automotive and healthcare customers, creating multiple multi billion dollar vertical markets. The Blackwell platform is in full production and forms the foundation for trillion parameter scale Generative AI. The combination of Grace CPU, Blackwell GPUs, NV Link, Quantum Spectrum, Nixon switches, high speed interconnects and a rich ecosystem of software and partners let us expand and offer a richer and more complete solution for AI factories than previous generations. Spectromax opens a brand new market for us to bring large scale AI to Ethernet only data centers and NVIDIA NEMS is our new software offering that delivers enterprise grade optimized generative AI to run on CUDA everywhere from the cloud to on Prem data centers to RTXAIPCS. Through our expansive network of ecosystem partners from Blackwell to Spectrum X to NEMS, we are poised for the next wave of growth. Thank you.
Simona Jankowski (Head of Investor Relations)
Thank you, Jensen. We will now open the call for questions. Operator, could you please pause for questions?
Regina (Operator)
At this time, I would like to remind everyone in order to ask a question, press *, then the number one on your telephone keypad. We'll pause for just a moment to compile the Q&A roster. As a reminder, please limit yourself to one question.
Regina (Operator)
Your first question comes from the line of Stacy Rascon with Bernstein. Please go ahead.
Jensen Huang (President and CEO)
Hi, guys. Thanks for taking my questions. My first.
Unknown (Sell-side Analyst)
One, I wanted to drill a little bit into the the Blackwell comments that it's in full production now. What does that suggest with regard to shipments and delivery timing? If that product is doesn't sound like it's sampling anymore, what what does it mean when when, when the reduction customers hands if it's in production now?
Jensen Huang (President and CEO)
We will be shipping well. We've been in production for for a little bit of time, but our production shipments will start in Q2 and ramp in Q3 and customers should have data centers stood up in Q4. Got it. So this this year we will see Blackwell revenue, sounds like we will see a lot of Blackwell revenue this year.
Regina (Operator)
Our next question will come from the line of Timothy R Curry with CBS. Please go ahead.
Jensen Huang (President and CEO)
Thanks a lot. I wanted to ask Jensen about the deployment of, you know, Blackwell versus Hopper. Just given the system's nature and, you know, all the demand for, you know, GB that you have, How does the deployment of this stuff differ from Hopper? I guess I asked because liquid cooling at scale hasn't been done before and there's some engineering challenges both at the node level and within the data center. So do these complexities sort of elongate the transition? And how do you sort of, you know, think about how that's all going? Thanks. Yeah, but Blackwell comes in many configurations. Blackwell is a platform, not a GPU. And the platform includes support for air cooled liquor cooled X86 and Grace InfiniBand. Now Spectrum X and very large MV link domain that I demonstrated at GTC, that I showed at GTC. And so for some customers, they will ramp into their existing install base of data centers that are already shipping hoppers. They will easily transition from H100 to H200 to B100. And so Blackwell systems have been designed to be backwards compatible, if you will, electrically, mechanically. And of course the software stack that runs on Hopper will run fantastically on on Blackwell. We also have been priming the pump, if you will, with the entire ecosystem, getting them ready for liquor cooling. We've been talking to the ecosystem a block about Blackwell for quite some time. And the CSPS, the data centers, the ODM, the system makers, our supply chain beyond them, the cooling, the cooling, supply chain base, liquid cooling, supply chain base, data center, supply chain base. No one is going to be surprised with Blackwell coming and the capabilities that we would like to deliver with Grace Blackwell 200 GB 200 is going to be exceptional.
Regina (Operator)
Our next question will come from the line of Vivek Arya with Bank of America Securities. Please go ahead.
Jensen Huang (President and CEO)
Thanks for taking my question. Jensen, how are you ensuring that there is enough utilization of your products and that there isn't a pull ahead or hoarding?
Unknown (Sell-side Analyst)
Behaviour because of tight supply, competition or other factors? What? What?
Jensen Huang (President and CEO)
Basically, what checks have you built in the system to give us confidence that monetization is keeping pace with your really, you know, very strong shipment growth? Well, I guess, I guess there's the there's the big picture view that I'll come to and then, but I'll, I'll answer your question directly. The demand for for our GP us in all the data centers is incredible. We're racing every single day. And the reason for that is because applications like ChatGPT and GPT 4 O and now it's going to be multi modality and Gemini and it's rampant anthropic and you know, all, all of the work that's being done at all the CSPS are consuming every GPU that's out there. There's also a long line of generative AI start-ups, some 15,020 thousand start-ups that in all different fields from from multimedia to digital characters, of course, all kinds of design tool, application, productivity applications, digital biology, the movement, the moving of the AV industry to video so that they can train end to end models to expand the operating domain of self driving cars. The list is just quite extraordinary. We're racing actually customers are, are, are putting a lot of pressure on us to deliver that, deliver the systems and stand it up as quickly as possible. And of course, I haven't even mentioned all of the sovereign AIS who would like to train all of their regional natural natural resource of their country, which is their data to train their, their regional models. And there's a lot of pressure to stand those systems up. So, so anyhow, the demand I think is, is really, really high and it's outstrips our supply longer term. That's what that's, you know, that's the reason why I jumped into to make a few comments long longer term, you know, we're, we're completely redesigning how computers work. And it's this is, this is a platform shift. Of course, it's been compared to other platform shifts in the past, but but time will will clearly tell that that this is much, much more profound than previous platform shifts. And the reason for that is because the computer is no longer an instruction driven only computer. It's an intention understanding computer and it understands it understand of course, the way we interact with it, but it also understands our meaning, what we intend that we asked it to do and has the ability to reason inference iteratively to to process a plan and come back with a solution. And so so every aspect of the computer is changing in such a way that instead of retrieving pre recorded files, it is now generating contextually relevant intelligent answers. And so that that's going to change computing stacks all over the world. And you saw a build that in fact, even the PC computing stack is going to get revolutionized. And this is just the beginning of all the things that, you know, what people see today are the beginning of the things that we're working in our labs and, and the things that we're doing with all the startups and large companies and developers all over the world. It's going to be, it's going to be quite, quite extraordinary.
Regina (Operator)
Our next question will come from the line of Joe Moore with Morgan Stanley. Please go ahead.
Jensen Huang (President and CEO)
Great. Thank you. Understand what you just said about how strong demand is.
Unknown (Sell-side Analyst)
You have a lot of demand for H200 and for Blackwell products. Do you anticipate any kind of pause with Hopper and H100 as you sort of migrate to those products? Will people wait for those new?
Jensen Huang (President and CEO)
Products which would be a good product.
Unknown (Sell-side Analyst)
To have or or do you think there's enough demand for H100 to sustain growth?
Jensen Huang (President and CEO)
We see increasing demand of Hopper through this quarter and we expect to be, we expect demand to outstrip supply for some time as we now transition to H200, as we transition to Blackwell, everybody is, is anxious to get their infrastructure online and the reason for that is because they're saving money and making money and they would like to do that as soon as possible.
Regina (Operator)
Our next question will come from the line of Toshiya Hari with Goldman Sachs. Please go ahead.
Jensen Huang (President and CEO)
Hi, thank you so much for taking the question. Jensen, I wanted to ask about competition. I think many of your cloud customers have announced, you know, new or or updates to to their existing internal programs, right. And parallel to to what they're working on with you guys, to what extent did you consider them as as competitors medium to long term? And in your view, do you think they're limited to addressing mostly internal workloads or could they be broader and and what they address going forward? Thank you. Yeah, we're different in several ways. First, Nvidia's accelerated computing architecture allows customers to process every aspect of their pipeline from unstructured data processing to prepare for training to structure data processing, data frame processing, like sequel to prepare for training to training to inference. And as I was mentioning in my remarks, that inference has really fundamentally changed. It's now generation. It's not trying to just detect the cat, which is, which was plenty hard in, in, in itself, but it has to generate every pixel of a cat. And so so the generation process is a fundamentally different processing architecture. And it's one of the reasons why Tensor RTLM was so L so well received. We improved the performance in using the same chips on our architecture by a factor of three. That kind of tells you something about the richness of our architecture and the richness of our software. So 1 you could use NVIDIA for, for everything from computer vision to image processing to computer graphics to, you know, all, all modalities of computing. And as the world is is now suffering from computing cost and computing. Energy inflation because general purpose computing has run its course. Accelerated computing is really the the sustainable way of going forward. Accelerated computing is how you're going to save money in computing is how you're going to save energy in computing. And so the versatility of our of our platform results in the lowest TCO for their data centers. Second, we're in every cloud. And so for developers that are, are looking for a platform to develop on, starting with NVIDIA is always a great choice. And we, we, we're on Prem, we're in the cloud, you know, we're in computers of, of any size and shape. We're, we're practically everywhere. And so that's the second reason. The, the third reason has to do with the fact that, that, you know, we build AI factories. And this is, this is becoming more and apparent to people that that AI is not a chip problem only. It starts of course with very good chips and we build a whole bunch of chips for AI factories. But it's a systems problem. In fact, even even AI is now a systems problem. It's not just one large language model. It's a complicate complex system of a whole bunch of large language models are that are working together. And so the fact that NVIDIA builds the system causes us to optimize all of our chips to work together as a system, to be able to have software that operates as a system and to be able to optimize across the system. And just to put it in perspective in simple numbers, you know, if you had a, if you had a $5 billion infrastructure and you improve the performance by a factor of 2, which we routinely do. You know, when you improve the infrastructure by factor 2, the value to you is $5 billion. All the chips in that data center doesn't pay for it. And so the value of it is really quite extraordinary. And this the reason why today performance matters. Everything you know, this is this is at a time when when the highest performance is also the lowest cost because the infrastructure cost of carrying all of these chips cost a lot of money. And it takes a lot of money to to fund the data center, to operate the data center, the pimple that goes along with it, the power that goes along with it, the real estate that goes along with it and all of it, all of it adds up. And so the highest performance is also the lowest TCO.
Regina (Operator)
Our next question will come from the line of Matt Ramsey with TD Cowan. Please go ahead.
Unknown (Sell-side Analyst)
Thank you very much. Good afternoon everyone. Jensen, I been in the data center industry my whole career. I've never seen the the velocity that you guys are introducing new platforms at the same combination of the performance jumps that you're getting. I mean, 5X and in training, some of the stuff you talked about at GTC up to 30X in in inference and that it's, it's amazing thing to watch. But it also creates an interesting juxtaposition where the, the, the current generation of product that your customers are spending billions of dollars on it. It it's going to be not as competitive with your new stuff very, very much more quickly than the depreciation cycles of, of that product. So I, I, I'd like you to, if you wouldn't mind speak a little bit about how you're seeing that situation evolve itself with customers as you move to Blackwell. They're going to have very large installed bases, obviously software compatible, but large installed bases of product that's not nearly as performance as as your new generation stuff. And and and be interesting to hear. What do you see happening with customers along that path? Thank you.
Jensen Huang (President and CEO)
Yeah, I really appreciate it. 333 points that I'd like to make. If you're, if you're 5% into the build out versus if you're 95% into the build out, you're going to feel very differently. And because you're only 5% into the build out anyhow, you know, you, you build, you build as fast as you can. And you know when Blackwell comes, it's going to be terrific. And then after Blackwell, as as you mentioned, we have, we have other black walls coming and then there's a short there's you know, we're on a one year rhythm as you as we've explained to the world. And we want our customers to see our road map for as far as they like. But they're they're they're early in their build out anyways. And so they have to just keep on building. OK. And so there's going to be a whole bunch of chips coming at them and they just got to keep on building and just, you know, if you will performance average your way into it. So that's that's the smart things, the smart thing to do. They need to make money today. They want to save money today. And I and time is really, really valuable to them. Let me give you an example of time being really valuable. Why this idea of standing up a data center instantaneously is so valuable and getting this thing called time to train is so valuable. The reason for that is because the the next, the next company who reaches the next major plateau gets to announce a groundbreaking AI and the the second one after that gets to announce something that's, you know, 0.3% better. And so the question is, do you want to be repeatedly the company delivering groundbreaking AI or the company, you know, delivering .3% better? And that's the reason why this, this race, as in all technology races, the race is so important. And, and, and you're seeing this race across multiple companies because this is so vital to have technology leadership, for companies to trust the, the leadership and want to build on your platform and know that the, the, the platform that they're building on is going to get better and better. And so leadership matters a great deal. Time to train matters a great deal. The difference between time to train that is, you know, three months earlier just to get it done in order to get time to train on the three months project, you know, getting started three months earlier is everything. And so what's the reason why we're standing up hopper systems like mad right now? Because the next plateau is just around the corner. And so, so that's the second reason that the first, the first comment that you made is really a great comment, which is, you know, how is it that we're doing? We're moving so fast and, and advancing so quickly because we have all the stats here. We literally build the entire data center and we can monitor everything, measure everything, optimize across everything. We know where all the bottlenecks are. We're not guessing about it. We're not putting up PowerPoint slides that look good. We're actually, you know, we also like our PowerPoint slides to look good. But, but we, we're delivering systems that perform at scale. And the reason why we know the performance scale is because we built it all here. Now we, one of the things that we do that's a bit of a miracle is that we build entire AI infrastructures here. But then we, we disaggregated and integrated into our customers data centers however they like. But we know how it's going to perform and we know where the bottlenecks are. We know where we need to optimize with them and we know where we have to help them improve their infrastructure to achieve the most performance. This deep intimate knowledge at the entire data center scale is fundamentally what sets us apart today. You know, we build every single chip from the ground up. We know exactly how processing is done across the entire system. And so we understand exactly how it's going to perform and how to get the most out of it with every single generation. And so I appreciate those are the three points.
Regina (Operator)
Your next question will come from the line of Mark Lapakos with Evercore ISI. Please go ahead.
Unknown (Sell-side Analyst)
Hi, thanks for taking my question. Jensen, in the past, you've made the observation that general purpose computing ecosystems typically dominated each computing era. And I believe the argument was that they could adapt to different workloads, get higher utilization Dr. cost compute cycle down. And this is a motivation for why you were driving to a general purpose GPU Kudi ecosystem system for accelerated computing. And if I mischaracterize that observation, please do let me know. So the question is, given that the workloads that are driving demand for your solutions are being driven by neural network training and inferencing, which which on the surface seemed like a limited number of workloads, then then it might also seem to lend themselves to to custom solutions. And so then the question is, is that does the general purpose computing framework become more at risk or is there enough variability or a rapid enough evolution on these workloads that that support that historical general purpose framework? Thank you.
Jensen Huang (President and CEO)
Yeah. And these accelerated computing is versatile, but I wouldn't call it general purpose. Like for example, we wouldn't be very good at running the spreadsheet, you know, that was really designed for general purpose computing. And so there's a, there's a, the, the, the, the control loop of an operating system code probably isn't, isn't fantastic for general purpose computing, not for our, not for our, not for accelerated computing. And so I would say that, that we're versatile and that's usually the way I describe it. There's a, there's a rich domain of applications that were able to accelerate over the years, but they all have a lot of commonalities, you know, maybe, maybe some deep differences, but commonalities, you know, they're all things that I can run in parallel. They're all high, heavily threaded. 5% of the code represents 99% of the runtime, for example. You know, those are all properties of accelerated computing. The versatility of our platform and the fact that we designed entire systems is the reason why over the course of the last 10 years or so, the number of start-ups that you guys have asked me about in these conference calls is fairly large. And every single one of them, because of the brittleness of their architecture. The moment, the moment generative AI came along or the moment the fusion models came along, the moment the next models, you know, the next models are coming along now. And, and now all of a sudden, look at this large language models with memory because the large language model needs to have memory so they can carry on a conversation with you, understand the context. All of a sudden, the versatility of the grace memory became super important. And so each one of these advances and generative AI and the advancement of AI really begs for not having a, a widget that's designed for one model, but to have something that is really good for this entire domain, properties of this entire domain, but obeys the the first principles of software. That software is going to continue to evolve. That software is going to keep getting better and bigger. We believe in the scaling of these models. There's a lot of reasons why we're going to scale by easily a million times in the coming few years for for good reasons and and we're looking forward to it and we're ready for it. And so the versatility of our platform is really quite key. And it's not if you're too, if you're too brittle and too specific, you might as well just build an FPGA or you build an ASIC or something like that. But that's hardly a computer.
Regina (Operator)
Our next question will come from the line of Blaine Curtis with Jeffries. Please go ahead.
Unknown (Sell-side Analyst)
Hey, thanks for taking my question. I.
Jensen Huang (President and CEO)
Actually kind of curious.
Unknown (Sell-side Analyst)
I mean, you're being supply constrained. How do you think about, I mean, you came out with, with a product for China age 20, I'm assuming there'd be a ton of demand for it, but obviously you're trying to serve your, your customers with, you know, the other hopper products. Just kind of curious how you're thinking about that in the second-half, you could elaborate, you know any impact what you're thinking for sales as well as gross margin?
Jensen Huang (President and CEO)
I didn't hear the questions. Something bleeped out.
Simona Jankowski (Head of Investor Relations)
Age 20 and how you're thinking about allocating supply between the different hopper.
Colette Kress (Executive Vice President and CFO)
Products.
Jensen Huang (President and CEO)
Well, you know, we just, we, we have customers that we honor and and we do our best for every customer. It is the case that that our business in China is substantially lower than the levels of the past and and it's a lot more competitive in China now because of the limitations on our technology. And and so those those those matters are true. However, you know, we continue to, to do our best to serve the customers and the markets there and, and to the best of the, our, our ability. We'll, we'll, we'll do our best, you know. And so, but that I think overall the comments that we made about demand outstripping supply is, is for the entire entire market and and particularly so for H200 and Blackwell towards the end of the year.
Regina (Operator)
Our next question will come from the line of Srini Pajuri with the Raymond James. Please go ahead.
Jensen Huang (President and CEO)
Thank you, Jensen. Actually more of a clarification on what you said, GP200 systems, it looks like there is a significant demand for systems. Historically, I think you've sold a lot of HEX boats and some GPU's and the systems business was relatively small. So I'm just curious, you know, why is it that now you are seeing such a strong demand for systems going forward? Is it just the TCO or is it something else or is it just the architecture? Thank you. Yeah, I appreciate that. In fact, the way we sell GB 200 is the same. We disaggregate all of the components that make sense and we integrate it into computer makers. We have 100 different computer sys configurations that are coming, coming this year for Blackwell. And and that is that is off the charts. Hopper, Hopper frankly had only half, but that's at its peak. You know, it started out with with way less than that even. And, and so you're going to see liquid cooled version, air cooled version, X86 versions, grace versions, you know, so on, so forth. And there's, there's a whole bunch of systems that are being designed and, and they're offered from all of our ecosystem of great partners. Nothing, nothing has really changed. Now, of course, the Blackwell platform is, has expanded our offering tremendously. I the, the integration of CPUs and the much more compressed density of computing liquid cooling is going to save data centers a lot of money in provisioning power and not to mention to be more energy efficient. And so, so it's, it's, it's a much better solution. It's more expansive, meaning that we, we offer a lot more components of a data center and everybody wins. You know, the data center gets much higher performance networking from networking switches networking of course. Next, we have Ethernet now so that we can bring NVIDIA AI to a large scale NVIDIA AI to customers who only operate on know only know how to operate Ethernet because of the, the, the ecosystem that they have and and so, so Blackwell is much more expensive. We have a lot more to offer our customers this time, this this time, this generation around.
Regina (Operator)
Our next question will come from the line of William Stein with Truest Securities. Please go ahead.
Unknown (Sell-side Analyst)
Great. Thanks for taking my question, Jensen. At some point NVIDIA decided that when there were, while there were reasonably good CPUs available for data center operations, your ARM based race CPU provided some real advantage that made that technology worth delivering to customers, perhaps related to cost or power consumption or technical synergies between Grace and Hopper, Grace and Blackwell. Can you address whether there could be a similar dynamic that might emerge on the client side whereby while there are very good solutions, you've highlighted that you know Intel and Anti are very good partners and deliver great products in X86, but there might be some, especially an emerging AI work with some advantage that NVIDIA can deliver that others have more of a challenge.
Jensen Huang (President and CEO)
Well, you, you mentioned, you mentioned some really good reasons. It is true that for many of the applications, our partnership with X86, our X86 partners are, are, are really terrific and we build excellent systems together. But Grace allows us to do something that isn't possible with the configuration, the system configuration. Today, the memory system between Grace and Hopper are coherent and connected the, the interconnect between the two chips. You know, calling it 2 chips is almost weird because it's like a super chip. The two of them are connected with this, with this interface that's like a terabytes per second, you know, it's off the charts. And the memory that's used by Grace is LPDDR. It's the first data center grade low power memory. And so we save a lot of power on every single node. And then finally, because because of the architecture, because we can create our own architecture with the entire system. Now we could create something that has really large MV link domain, which is vitally important to the next generation large language models for inferencing. And so you saw that GB 200 has a 72 node MV link domain that's like 72 Blackwells connected together into one giant GPU. And so we needed, we needed Grace Blackwells to be able to do that. And so there's there are architectural reasons there software programming reasons and then there's system reasons that that are essential for us to build them that way. And so if we see opportunities like that. You know, we'll explore, explore it and, and today, as you saw, as you saw at the, the build yesterday, which I thought was really excellent, Satya announced the next generation PCs, Copilot plus PC, which, which which runs fantastically on Nvidia's RTXGPUS that are, that are shipping in laptops. But you know, it also supports ARM beautifully. And so it opens up, it opens up opportunities for system innovation, even for even for PCs.
Regina (Operator)
Our last question comes from the line of CJ Muse with Cantor Fitzgerald. Please go ahead.
Jensen Huang (President and CEO)
Yeah, good afternoon. Thank you for taking the question. I guess Jensen, a bit of a longer term question. I know Blackwell hasn't even lost yet, but obviously investors are are forward-looking and amidst rising potential competition from GPUs and custom ASICS, how are you thinking about NVIDIA Spaces innovation at your million fold scaling of the last decade, truly impressive CUDA sparsity, precision, grace, coherent connectivity. When you look forward, what frictions need to be solved in the coming decade? And I guess maybe more importantly, what are you, what are you willing to share with us today? Well, I can announce that after Blackwell, there's another chip and, and we are on a one year rhythm. And so I and, and you can, you can also count that count on us having new networking technology on a very fast rhythm. We're announcing Spectrum X for Ethernet, but we're all in on Ethernet and we have a really exciting road map coming for Ethernet. We have a rich, rich, rich ecosystem of partners. Dell announced that they're taking Spectrum X to market. We have a rich ecosystem of Mark customers and partners who are going to announce taking our entire AI factory architecture to market. And so for companies that want the ultimate performance, we have InfiniBand computing fabric. InfiniBand is a computing fabric, Ethernet's a network. And InfiniBand over the years started out as a computing fabric, became a better and better network. Ethernet is a network. And with Spectrum X, we're going to make it a much better computing fabric. And we're committed, fully committed to all three links, MV link computing Fabric for a single, single computing domain to InfiniBand computing Fabric, to Ethernet networking computing Fabric. And so, so we're going to take all three of them forward at a very fast clip. And so you're going to see new switches coming, new NICs coming, new capability, new software stacks that run on all three of them, new CP us, new GP us new networking next new switches, a mountain of chips that are that are coming and all of it. The beautiful thing is all of it runs CUDA and all of it runs our entire software stack. So if you invest today on our software stack, you know, without doing anything at all, it's just going to get faster and faster and faster and faster. And if you invest in our, our, our architecture today without doing anything, it will get go to more and more clouds and more and more data centers and everything just runs. And so, so I, I think the, the, the pace of innovation that we're bringing will drive up the capability on the one hand and drive down the TCO on the other hand. And so we should be able to scale out with the NVIDIA architecture for this new era of computing and start this new industrial revolution where we manufacture not just software anymore, but we manufacture artificial intelligence tokens. And we're going to do that at scale. Thank you.
Regina (Operator)
That will conclude our question and answer session and our call for today. We thank you all for joining, and you may now disconnect.
Other transcripts: Q2 FY27 transcript · Q1 FY27 transcript · Q4 FY26 transcript · Q3 FY26 transcript · Q2 FY26 transcript · Q1 FY26 transcript
All NVDA earnings · Back to NVDA overview
Transcript produced by Signals.AI from the company's own earnings call audio; speech recognition can mishear a word. Not personalized investment advice.