NVIDIA B200 SXM Tensor Core GPU

Brand: NVIDIA | Category: GPUs

SKU: B200-SXM-180GB | Part #: B200-SXM-180GB | MPN: B200-SXM-180GB

Contact for Pricing — Request a Quote

The NVIDIA B200 SXM Tensor Core GPU requires careful power delivery planning: its 700W consumption and 12V-2x6 / 16-pin server-class power connector demand enterprise-grade PSU capacity and proper chassis thermal design. The SXM form factor positions this accelerator for high-density server deployments where compute density and memory bandwidth are non-negotiable. With 180 GB HBM3e memory and 14.4 TB/s memory bandwidth, the B200-SXM-180GB delivers the memory capacity and interconnect speed required for production AI and scientific computing workloads at scale.

Built on NVIDIA's Blackwell architecture, this GPU targets three core use cases: LLM training, inference, and HPC. The 20 TFLOPS TF32 tensor performance positions the B200 as a workhorse for organizations training and deploying large language models, running batch inference at scale, and executing high-performance computing simulations.

Omnixon Global supplies the NVIDIA B200 SXM Tensor Core GPU (part number B200-SXM-180GB) for immediate integration into your enterprise AI and HPC infrastructure. Our technical team can assist with specifications verification, server compatibility assessment, and logistics coordination. To request a quote, submit your requirements through Omnixon Global's RFQ portal or contact our B2B solutions team directly.