NVIDIA RTX Pro Blackwell GPUs

The NVIDIA Blackwell architecture delivers a 4x leap in generative AI performance and 2.5x faster ray tracing for professional visual computing workflows.

Built on the 4nm custom TSMC process, Blackwell-based RTX GPUs integrate 5th Gen Tensor Cores and 4th Gen RT Cores to accelerate complex enterprise tasks. These professional units feature up to 582 TFLOPS of compute power and high-speed GDDR7 memory (offering 1.5 TB/s bandwidth) to handle massive 3D datasets and real-time LLM inference. By leveraging dedicated Decompression Engines and multi-node scaling, Blackwell enables engineers to render cinematic frames and train local AI models with unprecedented efficiency: reducing energy consumption by up to 25x compared to previous generations.

https://www.nvidia.com/en-us/design-visualization/rtx-6000/