NVIDIA RTX PRO 6000 Blackwell Server Edition
Enterprise AI and Visual Computing
Built on the NVIDIA Blackwell architecture, the RTX PRO 6000 Blackwell Server Edition is designed to accelerate demanding enterprise workloads, including generative AI, AI inference, scientific computing, data analytics, 3D rendering and professional visual computing.
With 96 GB of high-bandwidth GDDR7 memory and Error-Correcting Code (ECC), the GPU supports memory-intensive applications and complex computing workloads. Its passive thermal solution is intended for compatible server environments engineered to provide the required airflow.
Key Features
• 96 GB GDDR7 memory with ECC
• 24,064 CUDA cores and 752 fifth-generation Tensor Cores
• 188 fourth-generation RT Cores
• Up to 4 PFLOPS of peak FP4 AI performance
• PCI Express 5.0 x16 interface
• Configurable power consumption up to 600 W
• Passive cooling for compatible server systems
• Multi-Instance GPU (MIG) support for up to four 24 GB instances
Compatibility
This GPU is intended for compatible server platforms with appropriate PCIe support, power delivery and chassis airflow. Verify system compatibility and power requirements before ordering.
Felix Bernard –
La NVIDIA RTX PRO 6000 Blackwell Server Edition offre des performances exceptionnelles pour les applications d’IA et le calcul professionnel. L’installation s’est déroulée sans difficulté et la carte répond parfaitement à nos attentes. Un excellent choix pour les applications exigeantes en centre de données.