NVIDIA HGX B300 NVL16 servers available now. Learn more
Unleash Blackwell architecture in your data center with NVIDIA RTX PRO™ 6000 Server Edition. Perfect for demanding AI visualization, digital twins, and 3D content creation workloads.
Blackwell Cloud offers customizable server configurations, featuring up to 8x NVIDIA RTX PRO™ 6000 Server Edition GPUs, manufactured by various OEMs.
The NVIDIA RTX PRO™ 6000 Server Edition is designed for large-scale deployment of professional graphics workloads, accelerated rendering, and real-time digital twin simulations. With 96GB of GDDR7 memory per GPU and up to 8 GPUs per server, these systems bring unprecedented power to visualization and LLM inference pipelines.
Run large-scale 3D simulations and collaborative environments in real-time with low latency.
Accelerate complex rendering pipelines with memory-rich, parallelized GPU configurations.
Seamlessly integrate AI into video editing, upscaling, motion capture, and real-time content generation pipelines.
| GPU Architecture | NVIDIA Blackwell Architecture |
| CUDA® parallel processing cores | 24,064 |
| NVIDIA Tensor Cores | 752 (fifth-generation) |
| Single-precision performance (FP32) | 120 TFLOPS |
| Peak FP4 AI PFLOPS | 4 PFLOPS |
| RT Core performance | 355 TFLOPS |
| GPU memory | 96 GB GDDR7 with ECC |
| Memory interface | 512-bit |
| Memory bandwidth | 1597 GB/s |
| Multi-Instance GPU (MIG) | Up to 4 MIGs @ 24GB |
| Power consumption | Up to 600W (Configurable) |
| Graphics bus | PCI Express 5.0 x16 |
| Power connector | 1x PCIe CEM5 16-pin |