Registration Strip Icon for charts Register for streaming realtime charts, analysis tools, and prices.

Nvidia Unveils Blackwell: The Next-Gen AI Architecture Powering the Future of Computing

Bruno T
Latest News
March 19 2024 6:35AM

Nvidia (NASDAQ:NVDA) unveiled its latest advancement in graphics processor architecture named Blackwell on Monday, marking a significant leap forward from its existing Hopper technology. This new development aims to enhance the performance and efficiency of artificial intelligence (AI) systems, such as OpenAI’s ChatGPT, cementing Nvidia’s role as a key player in the AI hardware market. With a market value of $2.195 trillion as of last Friday, Nvidia stands as the third-largest company globally, trailing only behind tech giants Apple (AAPL) and Microsoft (MSFT).

The announcement was made at Nvidia’s event in San Jose, California, which drew an estimated crowd of 20,000 attendees. Dubbed the GPU Technology Conference (GTC), or “The Woodstock of Artificial Intelligence” by Bank of America analyst Vivek Arya, the event highlighted Nvidia’s increasing prominence in the tech industry.

Nvidia’s Blackwell architecture promises a revolution in generative AI capabilities, allowing systems to analyze up to 10 trillion parameters simultaneously, with a 25-fold reduction in energy consumption and cost compared to its predecessors. During the event, Nvidia CEO and co-founder Jensen Huang hailed the new technology as the driving force behind a new industrial revolution.

The Blackwell architecture will power the NVIDIA GB200 Grace Blackwell, a high-performance computing solution comprising two Blackwell GPUs and a Grace processor, tailored for server platforms used in AI model training and execution.

Tech giants such as AWS (Amazon), Microsoft, Google (Alphabet), and Oracle are among the first to receive Nvidia’s cutting-edge technology. Additionally, companies like Cisco, Dell, Hewlett Packard Enterprise, Lenovo, and Supermicro have struck deals to offer servers equipped with Blackwell GPUs.

Nvidia’s launch comes at a time when AI technology is fueling unprecedented growth for the company, with record-breaking sales figures and a surge in market valuation, mirroring the broader tech industry’s upward trend. In the fourth quarter alone, Nvidia reported a 265% year-on-year revenue increase to $22.1 billion, while net income soared to $12.829 billion, a 491% rise.

The AI boom has also prompted Nvidia’s competitors to step up their game. AMD introduced its MI300 AI chip series in December, aiming to challenge Nvidia’s dominance, while Intel announced its Gaudi 3 line, targeting the AI accelerator market later this year.

Originally founded in 1993, Nvidia was best known for its graphic processing units (GPUs) popular among gamers. However, the 2000s saw a shift as researchers began leveraging GPUs for parallel processing in AI applications, a task at which they excelled over traditional CPUs. Recognizing this potential, Nvidia developed Cuda, a programming language for GPU-based applications, and shifted its focus towards high-performance chips for servers, further establishing its foothold in the AI hardware market.