NVIDIA HGX B300 AI Server
The NVIDIA HGX B300 AI Server is a next-generation AI computing platform built on the NVIDIA Blackwell Ultra architecture. Featuring 8 Blackwell GPUs with ultra-high-speed NVLink 5 interconnect and liquid cooling, it delivers unprecedented performance for large language model (LLM) training, generative AI, and HPC workloads.
A Supercomputer Purpose-Built for AI and HPC
The NVIDIA HGX™ platform provides the foundation for the world’s most powerful AI and high-performance computing (HPC) systems. Built on NVIDIA Blackwell and Hopper architecture GPUs, HGX delivers unprecedented scalability, efficiency, and performance for training and inference workloads.
Flexible Architecture for Every Data Center
NVIDIA HGX enables system builders to design a broad range of configurations—from compact 4-GPU servers to massive 8-GPU platforms interconnected with NVLink® and NVSwitch™—delivering unmatched compute density for modern AI factories.
Unmatched Scalability and Performance
With support for NVIDIA NVLink, NVSwitch, and Grace CPU Superchips, the HGX platform enables unified GPU memory and extreme bandwidth—empowering large-language-model (LLM) training, generative AI, simulation, and scientific computing at massive scale.
Key Features
- Built with NVIDIA Blackwell and Hopper GPUs
- Up to 8× SXM GPUs per system with NVLink 5.0
- Supports NVIDIA Grace CPU Superchips
- Industry-leading performance and energy efficiency
- Ready for NVIDIA AI Enterprise and DGX BasePOD™ architectures
- Optimized for generative AI, HPC, and cloud deployments
| MFG Number | HGX B300 |
|---|---|
| Condition | Item Condition : Brand New |
| Price | $300,000.00 |
| Show Product Will Be Available Soon on FE | No |
| Product Card Description | Experience data-center-class AI power with the NVIDIA HGX B300 AI Server. Powered by 8 NVIDIA Blackwell Ultra GPUs and 2.1 TB of HBM3e memory, it achieves up to 144 petaFLOPS of FP4 performance. Designed for hyperscale AI factories, this system combines liquid-cooled efficiency with NVLink 5 interconnect for seamless multi-GPU scalability. |
| Order Processing Guidelines | Order Processing Guidelines:
Inquiry First – Please reach out to our team to discuss your requirements before placing an order. |
