Powering the new era of computing with the NVIDIA GB200 Grace Blackwell Superchip. 72 Blackwell GPUs delivering 30x faster real-time trillion-parameter LLM inference.
The GB200 NVL72 connects 36 Grace CPUs and 72 Blackwell GPUs in a rack-scale, liquid-cooled design. It boasts a 72-GPU NVLink domain that acts as a single, massive GPU and delivers 15x faster inference and 3x faster training compared to DGX H100 systems.
72 NVIDIA Blackwell GPUs, 36 NVIDIA Grace CPUs, 2,592 Arm Neoverse V2 cores
192 GB HBM3e per B200, 8 TB/s per GPU, 130 TB/s NVLink, 576 TB/s total system
1,440 PFLOPS FP4, 720 PFLOPS FP8/FP6, 360 PFLOPS FP16/BF16, 180 PFLOPS TF32
Fifth-generation NVLink, 130 TB/s bandwidth, seamless GPU-to-GPU communication
15x faster inference vs DGX H100, 3x faster training, 2.5x performance per B200 vs H200
Liquid-cooled rack-scale design, single massive 72-GPU domain, exascale computing
Revolutionary innovations that redefine the boundaries of AI computing performance, delivering unprecedented capabilities for the most demanding AI workloads.
GB200 NVL72 introduces cutting-edge capabilities and a second-generation Transformer Engine, which enables FP4 AI. When coupled with fifth-generation NVIDIA NVLink, it delivers 30x faster real-time LLM inference performance for trillion-parameter language models.
GB200 NVL72 includes a faster second-generation Transformer Engine, featuring FP8 precision, enabling a remarkable 4x faster training for large language models at scale. This breakthrough is complemented by the fifth-generation NVLink.
Unlocking the full potential of exascale computing and trillion-parameter AI models requires swift, seamless communication between every GPU in a server cluster. The fifth generation of NVLink is a scale–up interconnect that unleashes accelerated performance.
Liquid-cooled GB200 NVL72 racks reduce a data center's carbon footprint and energy consumption. Liquid cooling increases compute density, reduces floor space, and facilitates high-bandwidth, low-latency GPU communication.
The NVIDIA Grace CPU is a breakthrough processor designed for modern data centers running AI, cloud, and HPC applications. It provides outstanding performance and memory bandwidth with 2x the energy efficiency of today's leading server processors.
Databases play critical roles in handling, processing, and analyzing large volumes of data for enterprises. GB200 takes advantage of high-bandwidth memory performance, NVLink-C2C, and dedicated decompression engines to speed up key database queries by 18x compared to CPU.
Polarise has achieved NVIDIA Preferred Partner status and is listed as official NVIDIA Cloud Service Provider (CSP), solidifying our position as a trusted leader in cloud innovation. This designation is reserved for select partners who operate large clusters built in coordination with NVIDIA, adhering to a tested and optimized reference architecture.
Let's discuss your specific requirements in a personal conversation. I'll help you find the perfect AI infrastructure solution for your organization.
