NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation

NVIDIA Blackwell

The NVIDIA Blackwell architecture is designed for environments where performance, scalability, and support for advanced AI and simulation workloads are critical.

RTX PRO 6000 Blackwell Max-Q offers:

• 24,064 CUDA cores for extreme parallel computing
• 3511 TOPS AI for generative AI and inference
• 110 TFLOPS FP32 for simulation and scientific computing

AI environments and large models

RTX PRO 6000 Blackwell Max-Q enables large language models and advanced AI pipelines to run locally, without the need for cloud infrastructure.

• LLM and generative AI
• advanced data analysis
• AI inference and model training
• image and video processing

Multi-Instance GPU (MIG)

Support for 4 × MIG enables the GPU to be split into independent instances, allowing multiple users to work simultaneously or run multiple AI environments.

This solution increases resource efficiency and optimises workloads in development and production environments.

Workstation scaling

The ability to configure systems with 1 to 4 GPUs allows you to significantly increase computing power depending on your project needs.

This approach enables you to build high-performance AI, simulation and visualisation environments on a single platform.

Professional integration

The FHFL Dual Slot format and 4 × DisplayPort 2.1 allow integration into advanced workstations and multi-monitor configurations.

Maximum power consumption of 300 W ensures high performance with optimal energy efficiency.