VCNRTX6000ADA-SB

• 48GB GDDR6 Memory on a 384-Bit bus, 20Gbps Memory Clock
• Ada Lovelace architecture, 4nm lithography, 18, 176 CUDA cores, 142 RT Cores
• 915M Hz Core clock, 2505M Hz Boost Clock
• 91.1 TFLOPs Peak FP32, 2.85 TFLOPs Peak FP64 Performance
• PCIe 4.0 x16 interface, Dual-Slot PCIe Form factor, Blower Cooling
• 4x Display Port 1.4, Up to 4 displays, 5120 X 2880 Maximum Digital Resolution
• 300W TDP, Single 16-Pin Power connector, 9.5in X 4.4in
• DirectX 12 Ultimate, OpenGL 4.6, OpenCL 3.0, Vulkan 1.3 support; Windows 10/11 64-Bit and Linux 64-Bit

The VCNRTX6000ADA-SB is PNY Technologies' orderable part for the NVIDIA RTX 6000 Ada Generation professional graphics card, built on the Ada Lovelace architecture at 4nm with 18, 176 CUDA cores and 142 RT cores. It carries 48GB of GDDR6 memory on a 384-bit bus at a 20Gbps memory clock, runs a 915M Hz core clock with a 2505M Hz boost clock, and connects over PCIe 4.0 x16. The card is a dual-slot blower design measuring 9.5in by 4.4in with a 300W TDP, drawing power through a single 16-pin connector, and NVIDIA recommends a 300W system PSU headroom for it. Display output is four Display Port 1.4 connectors supporting up to four displays at resolutions to 5120 x 2880.

48GB professional cards are specified where dataset or model size, not frame rate, is the limit: AI model training and inference, large-scale 3D rendering and ray tracing, real-time virtual production and broadcast graphics, CAD, CAE and simulation, medical imaging, and GPU-accelerated scientific computing. The blower cooler and dual-slot form factor are chosen deliberately so multiple cards can be stacked in a workstation or rackmount chassis with front-to-back airflow, which axial-fan consumer cards do not support. Buyers are workstation OEMs and system integrators, AI and visual-effects studios, engineering and research organizations, and datacenter operators building GPU compute nodes.