19. March 2025
Nvidia Unveils High-Performance Gpus For Ai Powerhouses
Nvidia has announced its new line of AI-accelerating GPUs, called RTX PRO Blackwell. These high-performance GPUs are designed to accelerate a wide range of applications, including AI training and inference, computer vision, and gaming.
The first product in this series is the RTX PRO 6000 Blackwell Server Edition, expected to be available soon from leading data center system partners such as Cisco, Dell Technologies, Hewlett Packard Enterprise, Lenovo, and Supermicro. Cloud service providers and GPU cloud providers including AWS, Google Cloud, Microsoft Azure, and CoreWeave will also offer instances powered by the RTX Pro 6000 Blackwell Server Edition later this year.
In addition to the server edition, Nvidia has announced several new GPU products for various markets, including laptops. The RTX PRO 5000, RTX PRO 4500, and RTX PRO 4000 Blackwell GPUs will be available in the summer from partners such as Boxx, Dell, HP, and Lenovo. Nvidia RTX PRO Blackwell laptop GPUs are also expected to arrive later this year from the same partners.
Nvidia has made several key announcements about its future GPU products, including a custom-designed CPU called Vera Rubin. This CPU is designed to power the company’s future GPUs and features 88 custom ARM cores with 176 threads connected to Rubin GPUs via a high-speed NVLink interface.
The centerpiece of Nvidia’s announcement was Vera Rubin itself, a GPU named after famous astronomer Vera Rubin. When configured in a full NVL144 rack, Vera Rubin delivers 3.6 exaflops of FP4 inference compute—three times more than Blackwell Ultra’s 1.1 exaflops in a similar rack configuration.
Vera Rubin features two GPUs together on one die that deliver 50 petaflops of FP4 inference performance per chip. The system is designed to support a wide range of applications, from AI training and inference to computer vision and gaming.
The Vera CPU is a key component of Nvidia’s future GPU products, featuring 88 custom ARM cores with 176 threads connected to Rubin GPUs via a high-speed 1.8 TB/s NVLink interface. While no release date has been announced, it is expected that Vera Rubin will be available in the second half of 2026.
Nvidia’s new RTX PRO Blackwell series and Vera Rubin represent significant advancements in AI acceleration and computing performance. They are designed to accelerate a wide range of applications, from AI training and inference to computer vision and gaming. These products mark an exciting milestone for Nvidia, and their potential impact on various industries cannot be overstated.
The significance of Nvidia’s announcement lies in the company’s commitment to accelerating AI-driven technologies. With the introduction of RTX PRO Blackwell and Vera Rubin, Nvidia is positioning itself as a leader in AI acceleration and computing performance. As these products become available, they are expected to have a major impact on various industries, from healthcare and finance to gaming and entertainment.
The potential applications of Nvidia’s new GPU products are vast and varied. In the field of AI training, RTX PRO Blackwell will enable developers to accelerate their models and improve their accuracy. In computer vision, the product will allow for faster and more accurate object detection, facial recognition, and image segmentation. Gaming will also benefit from the improved performance of Nvidia’s new GPUs.
In conclusion, Nvidia’s new line of AI-accelerating GPUs, including RTX PRO Blackwell and Vera Rubin, represent significant advancements in computing performance and AI acceleration. These products are designed to accelerate a wide range of applications and mark an exciting milestone for Nvidia. As these products become available, they will have a major impact on various industries and pave the way for future innovations in AI-driven technologies.