The PNY NVIDIA RTX PRO 6000 Blackwell Server Edition Graphics Card is a data center–class GPU engineered for AI, HPC, and enterprise workloads in Global. Built on the advanced NVIDIA Blackwell architecture, it delivers exceptional computational power, scalability, and reliability for demanding server environments.
Featuring 24,064 CUDA cores, 752 5th Gen Tensor Cores, and 188 4th Gen RT Cores, this GPU is optimized for AI training, inference, large-scale simulations, and real-time ray tracing. It is equipped with a massive 96GB of GDDR7 memory with ECC on a 512-bit memory interface, delivering up to 1.6 TB/s memory bandwidth. This ensures ultra-fast data processing and high reliability for mission-critical workloads such as deep learning, scientific computing, and complex data analysis.
Designed for enterprise scalability, the GPU supports Multi-Instance GPU (MIG) with up to 4 instances of 24GB each, allowing multiple workloads to run securely and efficiently on a single GPU. It also features confidential computing support and secure boot with root of trust, ensuring data protection and system integrity in enterprise environments.
The card uses a passive cooling solution, making it ideal for server deployments with controlled airflow. It operates within a configurable power range of 400W to 600W, providing flexibility based on performance requirements. Built on the PCI Express 5.0 x16 interface with 128 GB/s bidirectional bandwidth, it ensures fast communication between system components.
Connectivity includes 4 x DisplayPort 2.1 outputs, while advanced media engines with 4x NVENC, 4x NVDEC, and 4x JPEG support enable high-performance video processing workflows.
With its FHFL dual-slot form factor, 16-pin power connector, and enterprise-grade features, the PNY NVIDIA RTX PRO 6000 Blackwell Server Edition is the ultimate solution for AI infrastructure, cloud computing, and high-performance data centers in Global.
