New NVIDIA® RTX PRO™ Blackwell Workstation Edition GPUs

Groundbreaking performance in AI, ray tracing, and neural rendering technology to power the most innovative workflows in design, engineering, and beyond.

NVIDIA Logo

Built for Professionals
BOXX workstations powered by NVIDIA® RTX PRO™ Blackwell Generation Desktop GPUs


UNPARALLELED PERFORMANCE

NIVIDIA RTX PRO 6000 Blackwell

Designed for professionals who demand the best, NVIDIA RTX PRO™ Blackwell Edition Desktop GPUs inside BOXX NVIDIA-Certified Systems™ deliver unparalleled performance, reliability, and support. Each GPU is rigorously tested for a wide range of engineering, design, and AI workflows and optimized through enterprise drivers. BOXX workstations equipped with NVIDIA RTX PRO™ GPUs are available in multiple form factors (APEXX deskside systems and RAXX and FLEXX data center platforms) and ideal for enterprise deployments and mission-critical work. Liquid-cooled to optimize performance, BOXX APEXX small tower deskside workstations can accommodate up to two NVIDIA RTX PRO™ 4000, 4500, or 5000 Blackwell GPUs, while APEXX mid tower workstations can support up to four. *NVIDIA RTX PRO™ 6000 Blackwell Workstation Edition GPUs are available in APEXX mid tower workstations only.

 

Ready to elevate your workflow with BOXX workstations powered by NVIDIA RTX PRO GPUs? Talk to an expert today to find the perfect solution for your needs.


DESIGNED FOR PRO WORKFLOWS

  • NVIDIA Blackwell Architecture
  • 24,064 CUDA Parallel Processing Cores
  • 752 NVIDIA Tensor Cores
  • 188 NVIDIA RT Cores
  • 125 TFLOPS of single precision performance
  • 4000 AI TOPS
  • 96GB GDDR7 with EEC
The Blackwell Streaming Multiprocessor (SM) integrates next-gen CUDA cores and neural shaders to optimize AI-augmented graphics and compute performance. With up to 1.4X higher FP32 throughput over the previous generation, Blackwell SM architecture delivers up to 125 TFLOPS of single precision performance for:
Enable real-time AI rendering with 5th Gen Tensor Cores designed to power AI graphics, accelerating deep learning matrix operations critical to neural network training and inference. These cores offer 3X the AI processing power of the previous generation to accelerate:
  • Local LLM prototyping
  • Industrial scale referencing
  • Efficient deployment of next-gen AI models
4th Gen RT Cores enable true-to-life neural graphics by delivering up to 2X faster ray tracing performance over the previous generation, achieving photoreal scene rendering and physically accurate shadows, reflections and refractions. Engineered for industrial-grade precision, these cores optimize:
With up to 96GB of GPU memory, GDDR7 memory significantly boosts bandwidth and capacity, accelerating ray tracing, rendering, AI workloads and:
  • Content creation across all applications
  • Larger, more complex datasets
  • Massive 3D and AI projects, including immersive VR

ENHANCED VIDEO

Fast icon

NVIDIA Blackwell GPU architecture offers hardware encode and decode support for 4:2:2 chromasampled video, providing major performance benefits to the format so you’ll spend less time waiting and more time creating.

4k Vido Icon

9th-Gen NVIDIA NVENC video encode engines accelerate video encoding video encoding and improve quality for professional video applications

icon

6th-Gen NVDEC video decode engines double H.264 decoding throughput compared to the previous generation and optimize playback, video data ingestion, and AI-powered video editing features.


BLACKWELL MAX-Q

For extreme performance and power efficiency, BOXX workstations are also available with NVIDIA RTX PRO 6000 Max-Q Blackwell Workstation Edition GPUs. These powerful but smaller GPUs offer reduced power consumption (300W as compared to 600W in the standard NVIDIA RTX PRO 6000 Blackwell Workstation Edition GPU), while their smaller footprint maximizes efficiency in dense, multi-GPU configurations and/or small form factor systems.

To learn more about these revolutionary GPUs, read the BOXX and NVIDIA press releases.

 

APEXX3 with NIVIDIA RTX PRO Max-q GPU