NVIDIA, a leading technology company, has unveiled its groundbreaking Blackwell GPU architecture, marking a significant step towards accelerating computing capabilities and enabling organizations to develop and deploy real-time generative AI models with trillion-parameter large language models. Named after the renowned mathematician and statistician David Harold Blackwell, the Blackwell platform is poised to deliver up to 25 times lower cost and energy consumption in comparison to its predecessor, the Hopper architecture. Jensen Huang, the Founder and CEO of NVIDIA, emphasized the pivotal role of generative AI, stating, “Blackwell is the engine to power this new industrial revolution.”

The Blackwell architecture introduces six transformative technologies, including the world’s most powerful chip boasting 208 billion transistors, a second-generation Transformer Engine supporting double the compute and model sizes, fifth-generation NVLink interconnect for high-speed multi-GPU communication, and advanced engines for reliability, security, and data decompression. At the core of Blackwell lies the NVIDIA GB200 Grace Blackwell Superchip, a fusion of two B200 Tensor Core GPUs with a Grace CPU interconnected via an ultra-fast 900GB/s NVLink interconnect. Systems like the liquid-cooled GB200 NVL72 platform can combine multiple GB200 Superchips, offering an impressive 1.4 exaflops of AI performance with up to 72 Blackwell GPUs and 36 Grace CPUs.

Major cloud providers, including Amazon Web Services, Google Cloud, Microsoft Azure, and Oracle Cloud Infrastructure, have already pledged support to offer Blackwell-powered instances. Additionally, prominent partners such as Dell Technologies, Meta, Microsoft, OpenAI, Oracle, and Tesla, among others, are planning to incorporate Blackwell products into their offerings across hardware, software, and sovereign clouds. Sundar Pichai, CEO of Alphabet and Google, expressed enthusiasm about the partnership with NVIDIA, stating, “We are fortunate to have a longstanding partnership with NVIDIA, and look forward to bringing the breakthrough capabilities of the Blackwell GPU to our Cloud customers and teams across Google to accelerate future discoveries.”

The Blackwell architecture, coupled with its supporting software stack, is poised to drive breakthroughs across various industries, ranging from engineering and chip design to scientific computing and generative AI. Mark Zuckerberg, Founder and CEO of Meta, highlighted the importance of AI in powering various aspects of their operations, expressing eagerness to utilize NVIDIA’s Blackwell for training their open-source Llama models and developing the next generation of Meta AI and consumer products. With its remarkable performance gains and efficiency improvements, Blackwell holds the potential to realize the long-awaited goal of achieving real-time trillion-parameter AI applications for enterprises, ushering in a new era of AI innovation and advancements.