Unveiling Tomorrow’s Innovations: ZENMEV’s GTC Experience with NVIDIA
This content is provided by a sponsor
/ZENMEV/ – The much-anticipated event for anyone in the world of technology and innovation is the NVIDIA GTC 2024. Regaled as the IT event of the industry, the GPU Technology Conference (GTC) united developers, researchers, IT professionals, business leaders, and students in what is considered the forefront of exploring the transformative capabilities of Artificial Intelligence, accelerated computing, data science, and more.
ZENMEV was part of the NVIDIA GTC 2024, which took place from March 17th to 21st, 2024. This event, followed by the Al Conference and Expo, had NVIDIA CEO Jensen Huang deliver the keynote on March 18th.
What Is NVIDIA GTC?
Photo: NVIDIA
A global event, the NVIDIA GPU Technology Conference (GTC) is a prestigious conference attended by developers, engineers, researchers, inventors, and IT professionals. It provides a platform for collaboration and knowledge sharing.
Centered around artificial intelligence (AI), the conference covers various topics, including AI, computer graphics, data science, machine learning, and autonomous systems.
Established in 2009 in San Jose, California, GTC initially focused on leveraging GPUs to address computing challenges. Over the years, the conference has evolved to highlight the diverse applications of AI and deep learning, spanning industries such as self-driving cars, healthcare, high-performance computing, professional visualization, and training programs offered by the NVIDIA Deep Learning Institute (DLI).
What Did We See?
NVIDIA’s annual GTC event happened non-virtually in 2024, thanks to the pandemic being a thing of the past. NVIDIA wrapped up its 16th GPU Technology Conference (GTC) on 21 March 2024.
Two years after CEO Jensen Huang introduced the Hopper architecture-based GPUs at GTC 2022, their popularity among AI developers has made them the choice for AI training and inferencing. With its spot secured as the heart of the AI revolution, NVIDIA had a host of new offerings at GTC 2024.
As is customary, NVIDIA CEO and Founder Jensen Huang gave the keynote, followed by 900 inspiring sessions, over 300 exhibits, more than a technical workshop, and tons of unique networking events. GTC delivered something for every technical level and interest area.
What’s New?
With a 260% surge in the last twelve months, Nvidia’s stock has skyrocketed to the upper echelon of the world’s most valuable enterprises in 2024.
The same is seen in the company’s earnings, thanks to an ever-present demand for its graphics processors, which power artificial intelligence training and applications.
With the conference now, NVIDIA’s GTC developers conference held in San Jose, the company unveiled a series of products and platforms to strengthen its position as a leader in the AI market.
During the keynote address, Huang introduced Nvidia’s latest chip architecture, Blackwell, and associated products, such as the B200, the company’s newest AI chip.
Touted as the “Era of AI” by NVIDIA, given each week in the AI galore, this week saw:
The United Nations General Assembly adopted a global resolution on AI.
Discussions between Apple and Google centered on utilizing Google’s Gemini LLM for Apple’s devices.
Google DeepMind collaborated with the Liverpool Football Club to introduce TacticAI, an AI soccer assistant.
Stability AI launched Stable Video 3D (SV3D) for image-to-video conversion.
Sakana AI developed a merging technique based on evolutionary algorithms for basic model creation.
Google faced a penalty of €250 million for utilizing copyrighted material in its AI technology.
xAI made the Grok-1 LLM open-source.
With 64 billion parameters, Apple’s CEO announced ongoing work on MM1, a generative AI model.
And the most impressive part? NVIDIA’s chips power nearly all of this.
With the stock price shooting up, NVIDIA’s market capitalization reached over $2.28 trillion, making it one of the most valued companies in the world and Huang an important figure in tech.
What’s more, it’s not just about the hardware anymore. NVIDIA’s offerings are gradually turning it into a complete AI development powerhouse.
NVIDIA GTC 2024 Highlights
Blackwell GPUs
With the introduction of the new Blackwell GPUs offering raw compute gains for AI development, CEO Jensen Huang broke open the news in his keynote address. He says the flagship GB200 chip features two Blackwell NVIDIA B200 Tensor Core GPUs, a new Transformer Engine, and a GB200NVL 72 cluster for 30% higher performance. Having become available later this year, the Blackwell GPUs received more than their share of interest from major players like Microsoft and Amazon.
Scalable AI Infrastructure
NVIDIA announced the DGX SuperPOD, comprising multiple DGX GB200 systems, each with 36 NVIDIA GB200 Superchips. Delivering 11.5 exaflops of performance and 240 terabytes of memory, scalable up to eight systems, the DGX sets new standards. Moreover, Amazon Web Services upgraded Project Ceiba to include Blackwell GPUs, boosting its performance to 414 exaflops.
NVIDIA-MediaTek Partnership
NVIDIA licensed its GPU tech to MediaTek, resulting in four new chips for the automotive industry. The Dimensity Auto Cockpit includes chips with NVIDIA RTX GPU, Arm V9-A CPU cores, and multi-camera support. This partnership aims to bring AI processing directly to vehicles, eliminating the need for cloud connections.
Project GR00T
Project GR00T focuses on developing a general-purpose foundation model for humanoid robots. NVIDIA also unveiled Jetson Thor, a computer designed to run multimodal generative AI models based on Blackwell GPUs. This project aims to create an AI platform for leading roboticists worldwide.
API Availability
NVIDIA Omniverse Cloud is now available through five APIs, enabling integration into existing design and automation software applications. And APIs allow users to stream Omniverse OpenUSD files onto Apple Vision Pro, expanding its industrial use cases.
Bridging the AI Divide
NVIDIA Inference Microservices (NIM) were introduced as cloud-native microservices for generative AI deployment. This is believed to provide access to pre-built containers, models, APIs, and inference engines, reducing deployment time.
Partner Launches
Dell, Hewlett Packard Enterprise, Supermicro, and Lenovo announced generative AI infrastructure in collaboration with NVIDIA.
ClearML introduced Fractional GPU, a tool to manage NVIDIA GPUs and optimize GPU utilization.
Based on NVIDIA’s tech, Balbix’s BX4 AI Engine offers cyber risk management and security control measures.
Conclusion
Overall, the GTC event proved once again why it’s a flagship conference with attendees around the globe. Uniting the world of technology under its gigantic roof, NVIDIA’s 2024 conference was a resounding success, with its past and present ventures holding down its forte. ZENMEV has had an amazing experience being a part of the GTC event hosted by NVIDIA. And are excited to know what’s in store.
Disclaimer: This publication is sponsored. Coinspeaker does not endorse or assume responsibility for the content, accuracy, quality, advertising, products, or other materials on this web page. Readers are advised to conduct their own research before engaging with any company mentioned. Please note that the featured information is not intended as, and shall not be understood or construed as legal, tax, investment, financial, or other advice. Nothing contained on this web page constitutes a solicitation, recommendation, endorsement, or offer by Coinspeaker or any third party service provider to buy or sell any cryptoassets or other financial instruments. Crypto assets are a high-risk investment. You should consider whether you understand the possibility of losing money due to leverage. None of the material should be considered as investment advice. Coinspeaker shall not be held liable, directly or indirectly, for any damages or losses arising from the use or reliance on any content, goods, or services featured on this web page.
