NVIDIA DGX GB200 Blackwell 1440GB GPU
Driving the Future of AI Innovation
Artificial intelligence is revolutionizing industries by automating tasks, enhancing customer experiences, uncovering valuable insights, and driving innovation. Once a futuristic concept, AI is now a transformative force reshaping the way businesses operate. However, as AI workloads grow more sophisticated, they demand compute capacity that often exceeds what traditional enterprise infrastructure can provide. To unlock the full potential of AI, organizations require high-performance computing, storage, and networking solutions that are secure, scalable, and efficient.
Introducing the NVIDIA DGX™ B200, the newest breakthrough in the NVIDIA DGX platform. This next-generation AI solution is powered by NVIDIA Blackwell GPUs and advanced high-speed interconnects, redefining the possibilities of generative AI. Equipped with eight Blackwell GPUs, the DGX B200 delivers exceptional performance, featuring an impressive 1.4 terabytes (TB) of GPU memory and 64 terabytes per second (TB/s) of memory bandwidth. It is purpose-built to handle the most demanding enterprise AI workloads with unmatched efficiency.
With the NVIDIA DGX B200, enterprises can empower their data scientists and developers with a universal AI supercomputer. This accelerates innovation, reduces time to insight, and helps businesses fully harness the transformative potential of AI.
Contact us for pricing
Tech Specs & Customization
NVIDIA DGX GB200 Blackwell 1,440GB 4TB AI Supercomputer
GPU: 8x NVIDIA Blackwell GPUs
GPU Memory: 1,440GB total
Performance: 72 petaFLOPS for training, 144 petaFLOPS for inference
NVIDIA® NVSwitch™: 2x
System Power Usage: Approximately 14.3kW max
CPU: 2 Intel® Xeon® Platinum 8570 Processors, 112 Cores total, 2.1 GHz (Base), 4 GHz (Max Boost)
System Memory: Up to 4TB
Networking:
4x OSFP ports for NVIDIA ConnectX-7 VPI, up to 400Gb/s InfiniBand/Ethernet
2x dual-port QSFP112 NVIDIA BlueField-3 DPU, up to 400Gb/s InfiniBand/Ethernet
Management Network: 10Gb/s onboard NIC with RJ45, 100Gb/s dual-port ethernet NIC, Host BMC with RJ45
Storage:
OS: 2x 1.9TB NVMe M.2
Internal: 8x 3.84TB NVMe U.2
Software: NVIDIA AI Enterprise, NVIDIA Base Command, DGX OS / Ubuntu
Rack Units (RU): 10 RU
System Dimensions: Height: 17.5in, Width: 19.0in, Length: 35.3in
Operating Temperature: 5–30°C (41–86°F)
NVIDIA GB200 NVL72
Type: Grace Blackwell Superchip
Memory Clock: 8Gbps HBM3E
Memory Bus Width: 2x2x4096-bit
Memory Bandwidth: 2x8TB/sec
VRAM: 384GB (2x2x96GB)
FP4 Dense Tensor: 20 PFLOPS
INT8/FP8 Dense Tensor: 10 P(FL)OPS
FP16 Dense Tensor: 5 PFLOPS
TF32 Dense Tensor: 2.5 PFLOPS
FP64 Dense Tensor: 90 TFLOPS
Interconnects: 2x NVLink 5 (1800GB/sec) + 2x PCIe 6.0 (256GB/sec)
GPU: 2x “Blackwell GPU”
GPU Transistor Count: 416B (2x2x104B)
TDP: 2700W
Manufacturing Process: TSMC 4NP
Interface: Superchip
Architecture: Grace + Blackwell
Q4 2024 RELEASE. Inquire for more information, lead times and pricing details.