Jensen Huang

Using NVIDIA software to optimize your fine-tuning, Inference and deployment for the new Llama 3.1 1. Tailored AI Solutions: NVIDIA AI software enables enterprises to build custom generative AI models using their proprietary data. This ensures that the AI solutions are highly relevant and effective for specific business needs and industry applications. 2. End-to-End AI Services: With comprehensive services spanning data curation, synthetic data generation, fine-tuning, retrieval, guardrails, and evaluation, enterprises can seamlessly develop and deploy robust AI models with NVIDIA's advanced NeMo Retriever microservices. 3. Accelerated Adoption and Impact: Leading companies like Accenture, Aramco, AT&T, and Uber are already harnessing these powerful tools to drive innovation and efficiency, demonstrating the transformative potential of custom Llama 3.1 models in real-world scenarios.

21

1 Comment

https://lnkd.in/eXbCsAnD “He who controls the GPUs, controls the universe.” Access to GPUs, and in particular those made by Nvidia, the leading supplier, is vital for any company that wants to be taken seriously in #ai. Nvidia’s dominance has driven its market value to more than $2trn. 'GPUs do the computational heavy-lifting needed to train & operate large #ai models. Yet, oddly, this is not what they were designed for. The initials stand for “graphics processing unit”, because such chips were originally designed to process video-game graphics. It turned out that, fortunately for Nvidia, they could be repurposed for #ai workloads. 'Might it be better to design specialist ai chips from scratch? That is what many companies, small & large, are now doing in a bid to topple Nvidia. Dedicated ai chips promise to make building and running ai models faster, cheaper or both. Any firm that can mount a credible threat will have no shortage of customers, who dislike its lofty prices and limited supplies. 'GPUs contain several thousand processing engines, or “cores”, which let them run thousands of versions of the same simple task at the same time. Running #ai models similarly involves running lots of copies of the same task in parallel. Figuring out how to rewrite ai code to run on GPUs was one of the factors that triggered the current ai boom. 'Yet GPUs have their limitations, particularly when it comes to the speed with which data can be shuffled on and off them... When training very large #ai models, some GPu cores may be idle as much as half of the time. Andrew Feldman, the boss of Cerebras, a startup based in Sunnyvale, CA, likens it to the gridlock in a grocery store on the day before Thanksgiving. “Everybody’s in a queue, so there are blockages in the parking lot, there are blockages in the aisles, blockages at the checkout. That’s exactly what’s happening with a gpu.” 'Cerebras’s response is to put 900,000 cores, plus lots of memory, onto a single, enormous chip, to reduce the complexity of connecting up multiple chips & piping data between them. 'Groq, a startup from Mountain View, CA, is taking a different approach. Its ai chips, called language processing units (lpus), are optimised to run #llms particularly quickly. 'Yet another approach is that taken by MatX, also based in CA. GPUs contain features & circuitry that provide flexibility for graphics, but are not needed for llms. The firm is working on getting rid of such unnecessary cruft.

14

3 Comments

What is next for NVIDIA? Wow, Jensen Huang delivered an incredible keynote at #Computex , and I’m convinced Wall Street doesn’t fully grasp Nvidia's potential. With #Nvidia now surpassing #Apple in size and its stock more than tripling over the past year, it’s tempting to think this might be the peak. But here’s the secret to finding great stocks: understanding the company’s products, not just their profits. And after this keynote, I’m confident Nvidia will soon top the charts as the biggest company on Earth. Let me show you why. 🚀 Why Nvidia is Unstoppable: 1) New #Rubin #GPUs in 2026: Revolutionary GPUs that build on the success of #Hopper and #Blackwell, ensuring full architectural compatibility. 2) Nvidia Inference Microservices (#Nims): Simplifying AI model deployment by allowing parallel processing on GPUs, vastly outperforming traditional CPUs. 3) #Omniverse Platform: A game-changing 3D simulation environment enabling real-time collaboration across industries, benefiting from massive network effects. 🔍 Key Takeaways from Jensen’s Keynote: 1) End-to-End AI Computing Platform: Nvidia isn’t just a hardware company; their comprehensive stack integrates GPU, CPU, DPU, NVLink, and advanced networking solutions. 2) Software-Driven Growth: The CUDA ecosystem and acceleration libraries are enabling breakthroughs in fields like gene sequencing, physics simulations, and generative AI. 3) Architectural & Backward Compatibility: Ensuring that each new generation of GPUs enhances the entire ecosystem without requiring customers to start from scratch. 🌐 Future-Proofing Data Centers: 1) Spectrum X Ethernet Solutions: Challenging #AMD and Intel by connecting GPUs in vast data centers, preparing for the surge in generative AI demands. 2) AI Everywhere: As AI becomes integral to daily interactions, Nvidia’s infrastructure is poised to handle the immense computational needs. 🔧 Practical Innovations: 1) NIMS in Action: Packaging everything needed for AI models to run efficiently and securely, reducing complexity and operational costs for companies. 2) Omniverse’s Impact: Enabling digital twins for factories, simulating warehouse operations, and training self-driving technologies, creating unmatched collaborative opportunities. 🔮 Looking Ahead: 1) Market Cap Milestones: Nvidia already surpasses Apple and could outpace Microsoft by year-end. 2) Robust Roadmap: Clear hardware development through 2027 with continued expansion into new markets. 3) Long-Term Investment: This isn’t just a short-term play. Nvidia's integrated approach and relentless innovation make it a cornerstone investment for the future. With a clear hardware roadmap and expanding into new markets, Nvidia’s potential is enormous. Understanding the science behind the NVIDIA leadership is crucial. Thanks for reading, and let’s keep an eye on Nvidia’s exciting journey ahead of 1:10 stock split! #Nvidia #AI #Investing #TechInnovation #LongTermGrowth CAIBots Crypto Exponentials

2

🌟 Exploring the Intersection of Business and Technology re AI! 🌟 Before my decade-long focus on AI, connecting technology and business in various domains has always been my passion; This is why I enjoyed this conversation with David Luan - No BS, No Twitter exaggerations, No clickbate - Just a super well informed, super smart David sharing his insights from years of hands-on experience: #Startegy, #Software, #Hardware, #Platforms, #ModelSpace and everything in between - LMK what you think of his projections. In This Episode: 1. The Biggest Lessons from OpenAI and Google Brain: What did OpenAI realize that no one else did that allowed them to steal the show with ChatGPT? Why did it take 6 years after the introduction of transformers for ChatGPT to be released? What are 1-2 of David’s biggest lessons from his time leading teams at OpenAI and Google Brain? 2. Foundation Models: The Hard Truths: Why does David strongly disagree that the performance of foundation models is at a stage of diminishing returns? Why does David believe there will only be 5-7 foundation model providers? What will separate those who win vs those who do not? Does David believe we are seeing the commoditization of foundation models? How and when will we solve core problems of both reasoning and memory for foundation models? 3. Bunding vs Unbundling: Why Chips Are Coming for Models: Why does David believe that Jensen and Nvidia have to move into the model layer to sustain their competitive advantage? Why does David believe that the largest model providers have to make their own chips to make their business model sustainable? What does David believe is the future of the chip and infrastructure layer? 4. The Application Layer: Why Everyone Will Have an Agent: What is the difference between traditional RPA vs agents? Why are agents a 1,000x larger business than RPA? In a world where everyone has an agent, what does the future of work look like? Why does David disagree with the notion of “selling the work” and not the tool? What is the business model for the next generation of application-layer AI companies? https://lnkd.in/dUpWWzvN

4

At #CES2025, NVIDIA CEO Jensen Huang announced a cornucopia of product innovation. I picked six to write about - check out my SiliconANGLE & theCUBE post that looks at 1) Cosmos 2) Omniverse 3) Blueprints for developers 4) RTX50 5) Project DIGITS 6) Enabling agentic AI https://lnkd.in/g6xbm4ns

30

2 Comments

The NVIDIA Blackwell platform and an optimized NV software stack boosted the limits of ample model training, setting new industry benchmarks in MLPerf Training 4.1, which is 2.2x faster than Hopper for Llama 2 and GPT-3. ✅ Up to 2.2x More Performance per GPU: Blackwell’s efficiency shines on large language models like Llama 2 and GPT-3, with groundbreaking per-GPU compute throughput and expanded HBM3e memory. ✅ Resource Efficiency: Thanks to new Tensor Core-optimized kernels, Blackwell needed only 64 GPUs to achieve results on GPT-3 benchmarks that previously required 256. ✅ Relentless Optimization: Consistent improvements on both Blackwell and Hopper architectures—enhanced performance and increased scale on GPT-3, and a 26% boost in fine-tuning for Llama 2. These achievements drive more powerful computing for partners and customers worldwide. MLPerf benchmarks showcase the potential for AI at scale, enabling informed investment decisions across industries. 🔗 Learn more about our MLPerf journey and the future of accelerated computing! #AI #MLPerf #Blackwell #NVIDIA

10

‘I Hope Suffering Happens To You’: Nvidia CEO’s Case For Hiring For Grit Over Pedigree by Jack Kelly Forbes “I wish you ample doses of pain and suffering.” That is the unconventional message NVIDIA CEO and founder Jensen Huang had for students at Stanford University, his alma mater, where he spoke last month at the Stanford Institute for Economic Policy Research. According to Huang, elite educational pedigrees alone are not necessarily the best predictor of success. “Greatness is not intelligence. Greatness comes from character. And character isn’t formed out of smart people, it’s formed out of people who suffered,” said the Stanford alum. Instead, the man who presides over one the world’s most valuable companies in terms of stock market valuation, now looks to find certain traits in job applicants that include resilience, grit and determination, which he contends lead to innovation and success in the workplace. Counterintuitively, Huang believes that having very high expectations coming out of an Ivy League education can potentially set people up for less resilience when faced with failure or adversity, which are inevitable in any career path. “Unfortunately, resilience matters in success,” he said to the students. “I don’t know how to teach it to you except for I hope suffering happens to you.” His comments indicate a view that real-world experience, involving struggles, obstacles and suffering, can instill the resilience needed for long-term success. Therefore, he believes that employers should look beyond just the prestige factor of a university and more closely at the inner drive, mindset and resolution of potential hires. “To this day I use the phrase ‘pain and suffering’ inside our company with great glee,” said the chief executive. “I mean that in a happy way, because you want to refine the character of your company. You want greatness out of them.” Hiring PHDs The Nvidia founder’s statements about valuing resilience over educational pedigrees align with the old Wall Street adage about hiring what they call “PHDs”—the poor, hungry and driven. The core parallels are prioritizing hunger, grit and motivation over academic credentials alone, believing that candidates who have had to overcome adversity, financial hardships or personal trauma offer invaluable drive and tenacity. The PHD concept essentially argues that struggle itself, if handled constructively, can be great preparation for thriving in ultra-competitive business environments where stamina and an insatiable work ethic are keys to success. A person who had to fight for opportunities may have a stronger inner drive and motivation to succeed compared to ... https://lnkd.in/e5XZitzS

4

2 Comments

Accelerated computing for multi-label classification unlocked. 🔢 ⚡ Discover how RAPIDS cuML seamlessly integrates with scikit-learn, bringing GPU performance to large and complex datasets.

15

Check out how the world’s first #generativeAI models for #OpenUSD development will be available as NVIDIA NIM microservices.

9

Worth thinking about! 🧠 The point is that most domain-specific problem solving, where the outcome or output is the result of some form of (quasi) optimal decision making, can be done very successfully by a suitable AI system (not necessarily LLM). And that applies to a lot of what companies do, across most businesses functions.

4

1 Comment

Buying a $1000+ USD GPU without a GPU holder is like buying an artwork without a frame. GPU sag is a relatively modern problem that GPUs are facing due to their size and weight having doubled in the past decade. This increased size and weight cause tremendous stress on the PCI-e slot, risking the PCB to break. Our Universal Vertical GPU Holder Kit is our #CoolerMaster solution for this problem, doubling down to provide a comprehensive solution. It totally eliminates the risk of #GPUsag and PCB damage while bringing a fresh visual flair to your #GamingPC. The installation is simple and tool-free - just watch our video to learn how: https://lnkd.in/e2TZ82Bg

10

🚀 NVIDIA NIM on Bedrock Marketplace 🚀 We’re thrilled to announce that NVIDIA NIM are now available on Amazon Bedrock Marketplace, SageMaker JumpStart, and AWS Marketplace. With this expansion, developers can now deploy high-performance, NVIDIA-optimized inference solutions for generative AI models across key AWS services, delivering: ✅ Faster AI inference ✅ Lower latency ✅ Cost efficiency From Meta’s Llama 3 to NVIDIA Nemotron and more, over 100 prebuilt NIM microservices are now at your fingertips. Tailored for scalability and secure deployment, NIM brings unmatched performance to models running on EC2, EKS, SageMaker, and Amazon Bedrock. This collaboration between NVIDIA and AWS accelerates innovation across industries, from healthcare to manufacturing, ensuring enterprises can meet the growing demand for generative AI applications. 💡 Ready to unlock the power of NVIDIA NIM on AWS? Explore the catalog, request a developer license, or start a 90-day NVIDIA AI Enterprise trial to experience the difference firsthand. Another great project by this wonderful team! Abhishek Sawarkar Jiahong Liu Kshitiz Gupta Timothy Ma Saurabh Trikande Charlie Huang Rohil Bhargava Let’s power the next wave of AI together! #GenerativeAI #NVIDIA #AWS #Bedrock #AIInference #AWSreInvent2024

44

1 Comment

Jensen Huang admits that Gen AI "rendering" game graphics from scratch is utter nonsense. He instead suggests that Nvidia GPU's fill in extra frames for a smoother experience, basically expanded DLSS. There's your AI rendering for games. Nvidia DLSS & AMD FSR. Great. AI getting framerate to 60fps but maybe with graphical glitches and hallucinations in those extra frames that aren't tied to player input. Is this really useful when the GPU can just kick out fully rendered, non-AI frames faster? I mean that is LITERALLY what the GPU does. ----- PC World: "Nvidia CEO: PC games will never be entirely rendered by AI Nvidia and AMD disagree on this key issue." (Mark Hachman) (Jan 7, 2025) "A key one: In a world where AI is increasingly used to generate or interpolate frames, is the end result a world in which PC graphics is entirely AI generated? No, Huang replied. There’s a reason we asked Huang the question. Nvidia says that while DLSS 3 could inject AI-generated frames between every GPU-rendered frame, DLSS 4 can infer three full frames off of a single traditional frame, as Brad Chacos noted in our earlier report on the GeForce 50-series reveal. A day earlier, rival AMD was essentially asked the same question. “Can I tell you that in the future, every pixel is going to be ML [machine learning]-generated? Yes, absolutely. It will be in the future,” AMD’s Frank Azor, its chief architect of gaming solutions, replied. Huang disagreed. “No, he replied to the question, asked by PCWorld’s Adam Patrick Murray. “The reason for that is because just remember when ChatGPT first came out, and we said, Oh, now let’s just generate the book. But nobody currently expects that. “And the reason for that is because you need to give it credit,” Huang continued. “You need to give it — it’s called condition. We now condition the chat or the prompts with context. Before you can answer a question, you have to understand the process. The context could be PDF, the context could be a web search. The context could be you told it exactly what the context is, right? “And so the same thing goes with video games. You have to give a context. And the context for video games has to not only be story-wise relevant, but it has to be spatial, world, spatially relevant. And so the way you condition, the way you give it context, is you give it some early pieces of geometry, or early pieces of textures, and it could generate, it could up the rest.” “In the future, 3D graphics would be 3D grounded condition generation,” he said. In DLSS 4, Nvidia’s GPU rasterization engine only renders one of out of the four forward-looking frames, Huang said. “And so out of four frames, 33 million pixels, we only rendered two [million]. Isn’t that a miracle?” The key, Hunag said, is that they have to be rendered precisely: “precisely the right ones, and from that conditioning we can generate the others.” (keep reading) PC World https://lnkd.in/gx-WP-ij #gameindustry #nvidia #ces2025 #amd

34

24 Comments