NVIDIA® GB200 NVL72
Introducing the cutting-edge NVIDIA GB200 compute tray, leveraging the full potential of NVIDIA’s innovative MGX design.
Introducing the cutting-edge NVIDIA GB200 compute tray, leveraging the full potential of NVIDIA’s innovative MGX design.
GB200 NVL72 is a rack-scale, liquid-cooled solution connecting 36 Grace CPUs and 72 Blackwell GPUs, enabling a single 72-GPU NVLink domain that delivers 30X faster real-time trillion-parameter LLM inference.
With GB200 SXM you get:
NVIDIA Blackwell architecture sets a new benchmark for accelerated computing with unparalleled performance, efficiency, and scalability, featuring 208 billion transistors on a custom TSMC 4NP process.
NVIDIA Grace CPU revolutionizes data centre computing with outstanding performance and memory bandwidth, offering 2X energy efficiency and unprecedented speed for AI, cloud, and HPC applications.
The fifth-generation NVIDIA NVLink unlocks exascale computing and trillion-parameter AI models, enabling swift and seamless communication between every GPU in your server cluster for accelerated performance.
NVIDIA Quantum-X800 InfiniBand, NVIDIA Spectrum X800 Ethernet, and NVIDIA BlueField®-3 DPUs enable efficient scalability across hundreds and thousands of Blackwell GPUs for optimal application performance.
| Form Factor | GB200 NVL72 |
|---|---|
| Configuration | 36 Grace CPU : 72 Blackwell GPUs |
| FP4 Tensor Core | 1,440 PFLOPS |
| FP8/FP6 Tensor Core | 720 PFLOPS |
| INT8 Tensor Core | 720 POPS |
| FP16/BF16 Tensor Core | 360 PFLOPS |
| TF32 Tensor Core | 180 PFLOPS |
| FP32 | 6,480 TFLOPS |
| FP64 | 3,240 TFLOPS |
| FP64 Tensor Core | 3,240 TFLOPS |
| GPU Memory | Bandwidth | Up to 13.5 TB HBM3e | 576 TB/s |
| NVLink Bandwidth | 130TB/s |
| CPU Core Count | 2,592 Arm® Neoverse V2 cores |
| CPU Memory | Bandwidth | Up to 17 TB LPDDR5X | Up to 18.4 TB/s |