Nvidia launches compact RTX Pro 4000 SFF and Pro 2000 GPUs with Blackwell but a low TDP.
Nvidia expands its GPU lineup with two new models based on the Blackwell architecture: the RTX Pro 4000 SFF Edition and the RTX Pro 2000. Both cards are compact and energy-efficient, yet according to the manufacturer, they offer more computing power for applications in engineering, 3D visualization, and content creation.
Nvidia launched Blackwell for workstations in March but at that time focused on powerful chips with TDPs up to 600 watts. These new models keep it to 70 watts.
Specifications
The RTX Pro 4000 SFF Edition and RTX Pro 2000 combine fourth-generation RT Cores with fifth-generation Tensor Cores in a casing that is half the size of a standard GPU. The new models target workstations with space or power consumption constraints. As usual, Nvidia refers to AI workloads as the goal of the Blackwell cards.
According to Nvidia, the RTX Pro 4000 SFF delivers up to 2.5 times higher AI performance, 1.7 times higher ray tracing performance, and 1.5 times more memory bandwidth than the previous generation, with a maximum TDP of 70 watts.
The RTX Pro 2000 is aimed at design and AI workflows for the broader audience and is expected to be up to 1.6 times faster in 3D modeling, 1.4 times faster in CAD, and 1.6 times faster in rendering. Additionally, Nvidia claims a 1.4 times higher speed in image generation and a 2.3 times faster processing of (AI) text generation.
Availability
Both GPUs will be available later this year. The RTX Pro 2000 will be available through PNY and TD SYNNEX and through system builders such as Dell Technologies, HP, and Lenovo. The RTX Pro 4000 SFF Edition will be available through global distribution partners and manufacturers like Dell Technologies, HP, and Lenovo.