NVIDIA officially launched on September 13 the RTX PRO 5500 Blackwell professional graphics card targeting professional workstations, edge computing and mid-scale inference scenarios. The card features the same core count as the consumer flagship RTX 5090 but comes with 96GB of ECC memory, approximately 40% improvement in double-precision floating-point performance, and a stricter power-management scheme, mainly targeting computer-aided design, building information modeling, complex scientific computing and mid-scale AI inference workloads. This is the first time NVIDIA has adopted the same core count across its professional and consumer product lines, and is seen by the industry as a strategic signal of NVIDIA further squeezing its discrete-graphics competitors in the professional graphics market.

Event Details

NVIDIA officially unveiled the RTX PRO 5500 Blackwell at a professional graphics product launch event on September 13. The product uses a Blackwell-architecture GPU core with the same core count as the consumer flagship RTX 5090, but is equipped with 96GB of ECC memory, approximately 40% improvement in double-precision floating-point performance over the previous-generation RTX PRO 5500 Ada, and a stricter power-management scheme to meet the stability requirements of professional workstations. The product is mainly targeted at computer-aided design, building information modeling, complex scientific computing and mid-scale AI inference scenarios. Pricing has not been announced, and shipments to professional graphics partners are expected to begin in mid-October. NVIDIA’s adoption of the same core count across its professional and consumer product lines for the first time is being read by the industry as a strategic move by NVIDIA to further integrate its product lines and increase pressure on its discrete-graphics competitors.

Edited by: GoodInfo Global News Desk