NVIDIA RTX Pro 6000 96GB GDDR7 Blackwell Server Graphic Card

NVIDIA RTX PRO 6000 Blackwell utilizes advanced GPU technology to deliver enhanced AI and graphics performance. This high-performing server graphics card has 752 fifth-gen Tensor Cores that offer improved FP4 and data precision, enabling users to run generative AI and LLM inference for AI-assisted professional workloads.

The NVIDIA RTX PRO 6000 96GB GDDR7 has a memory capacity of 96GB, with a 512-bit bus width and 1,597 GB/s bandwidth, facilitating adequate memory capacity for modeling AI models, datasets, and complex professional workloads while reducing memory errors.

With 24,064 CUDA cores, 752 fifth-gen Tensor Cores, and 188 fourth-gen RT cores, the NVIDIA RTX PRO 6000 Blackwell server edition can accelerate diverse workloads, including scientific computing, 3D rendering, engineering software, simulation, and AI.

The NVIDIA RTX PRO 6000 Blackwell server edition has a server-ready design that allows businesses to leverage air-cooled dual-slot or liquid-cooled single-slot nodes, PCIe Gen5 connectivity, and up to four isolated MIG instances, enabling flexible GPU resource allocation across diverse enterprise workloads.

NVIDIA RTX PRO 6000 Blackwell server edition has 96GB of GDDR7 memory with fifth-gen Tensor Cores to deliver increased memory capacity to support LLM inferencing, generative AI, computer vision, and other enterprise AI workloads.

Using the fourth-gen RT cores, the NVIDIA RTX PRO 6000 Blackwell can accelerate ray-traced graphics for 3D rendering, engineering visualization, and manufacturing or digital-twin applications. The large GPU memory also enables the handling of large and complex 3D models and visualizations.

The CUDA cores enable efficient scientific computing and data analytics acceleration. The graphics card also has 2 x ninth-gen NVENC and 2 x sixth-gen NVDEC to encode and decode professional videos with improved performance. Moreover, the PCIe Gen5 supports multi-GPU server configurations for high-performance computing.