NVIDIA has unveiled its Blackwell GPU architecture, targeting advanced AI and data center workloads. Announced at GTC 2024, Blackwell brings significant improvements in efficiency and performance, with availability expected later this year. Learn more about how this development could impact enterprise and research computing.
NVIDIA has officially introduced its new Blackwell GPU architecture, designed to power next-generation data centers and accelerate artificial intelligence workloads. The announcement was made during NVIDIA's GTC 2024 conference, where the company detailed the architecture's technical advancements and performance improvements.
The Blackwell platform features enhanced energy efficiency, increased memory bandwidth, and advanced security features tailored for large-scale AI training and inference. NVIDIA highlighted that the architecture is engineered to support demanding enterprise and research applications, including generative AI and high-performance computing.
Initial availability of Blackwell-based products is expected later in 2024, with leading cloud service providers and enterprise customers planning to integrate the new GPUs into their infrastructure. More information can be found on the official NVIDIA Blackwell page.