Nvidia launches RTX 4000 and 5000 workstation GPUs alongside new datacenter-specific accelerator for enterprise AI deployment.
Nvidia has introduced four new professional GPU products that complete the transition of its ProViz family to the Ada Lovelace architecture. The company has added three high-performance professional graphics cards for workstations as well as a server-grade professional board for remote graphics and light AI applications.
To address performance-demanding professional graphics applications such as computer aided design and digital content creation, Nvidia is rolling out the RTX 4000 20GB ($1,250, available September), the RTX 4500 24GB ($2,250, available October), and the RTX 5000 32GB ($4,000, available now), all based on the Ada Lovelace architecture. The company is also releasing the L40S datacenter board with 48GB of memory, available in the fall.
The RTX 4000 20GB largely mimics the RTX 4000 SFF product released several months prior, but uses a full-height PCB with a single-slot cooling system rated for 130W. It is powered by the AD104 GPU with 6,144 CUDA cores clocked higher than the SFF variant, delivering up to 26.7 FP32 TFLOPS of compute throughput, comparable to Nvidia's GeForce RTX 4070. The RTX 4500 24GB features the AD104 GPU with 7,680 CUDA cores offering up to 39.6 FP32 TFLOPS of compute performance, on par with the GeForce RTX 4070 Ti, and is equipped with a dual-slot cooling system with a blower fan rated for up to 210W of power.
The RTX 5000 32GB is based on a severely cut-down AD102 graphics processor with 12,800 CUDA cores that delivers compute performance of 65.3 FP32 TFLOPS, positioned to sit below the flagship RTX 6000 Ada ($6,800). The L40S Ada datacenter card uses the AD102 GPU with 18,176 CUDA cores that delivers 91.6 FP32 TFLOPS, in line with performance of the RTX 6000 Ada. While featuring passive cooling, the L40S Ada includes display outputs and can be installed into a workstation with sufficient airflow or a special blower attached to the board.
The new workstation boards will be used by companies including Boxx, Dell, HP, Lenovo, and Lambda in their upcoming workstations this fall, and will be available from Nvidia's resellers Arrow and Ingram, as well as from AIB suppliers Leadtek, PNY, and Ryoyo. According to Bob Pette, vice president of professional visualization at NVIDIA, "As generative AI transforms every industry, enterprises are increasingly seeking large-scale compute resources in the data center. OVX systems with NVIDIA L40S GPUs accelerate AI, graphics and video processing workloads, and meet the demanding performance requirements of an ever-increasing set of complex and diverse applications."