Power the AI factory of the future with the RTX Pro 6000 Blackwell Server Edition.
NVIDIA RTX PRO 6000 Blackwell Server Edition
The NVIDIA RTX Pro 6000 Server Edition redefines data center performance with next-generation AI and visual computing power. Featuring the groundbreaking NVIDIA Blackwell architecture, 24,064 CUDA cores, and 96GB of ultra-fast GDDR7 memory, it delivers extraordinary speed and efficiency for multimodal AI, rendering, and simulation. Purpose-built for scalable, always-on enterprise environments, it transforms your servers into powerful engines for innovation, intelligence, and discovery.
Key Features
Universal AI Acceleration
- Built on NVIDIA Blackwell architecture, the RTX Pro 6000 Server Edition delivers up to 5x faster AI performance with 5th-gen Tensor Cores and FP4 precision for next-generation inference, generative AI, and multimodal workflows.
Scalable, Data Center-Ready Design
- With 96GB of GDDR7 memory, 1.6 TB/s bandwidth, and PCIe Gen 5, it powers high-density server configurations for enterprise-grade AI, visualisation, and simulation at scale.
Enterprise Reliability, 24/7 Performance
- Engineered for continuous operation with a passive thermal design and NVIDIA-Certified server compatibility, it ensures peak uptime, security, and manageability across mission-critical environments.
Specifications
GPU Architecture
NVIDIA Blackwell
CUDA Cores
24,064
Tensor Cores
752 (5th Generation)
Ray Tracing Cores
188 (4th Generation)
AI Performance (FP4 Tensor Core)
Up to 4 PFLOPS (peak)
Single-Precision Performance (FP32)
120 TFLOPS
RT Core Performance
355 TFLOPS
GPU Memory | Bandwidth
96GB GDDR7 with ECC | 1.6 TB/s
Memory Interface
512-bit
Multi-Instance GPU (MIG)
Up to 4 instances @ 24GB each
Video Engines
4 x NVENC | 4 x NVDEC | 4 x NVJPEG (9th/6th Generation)
