Nvidia Distribution Australia
SKU:
NVIDIA GB300 (Gigabyte) DGX Station
NVIDIA GB300 (Gigabyte) DGX Station
The GIGABYTE W775-V10-L01 is a DGX Station-class deskside AI supercomputer powered by the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip. Designed for AI developers, researchers and data scientists, it supports large-scale model training, fine-tuning and inference directly from the desktop.
The system supports AI models with up to one trillion parameters and provides data-centre-class performance in a deskside format.
Click here to contact our dedicated AI sales team about the DGX Station
PDF Data sheet can be found here.
Click here to contact our dedicated AI sales team about the DGX Station
PDF Data sheet can be found here.
NVIDIA GB300 GIGABYTE DGX Station, Key Features
- NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip, combining a 72-core NVIDIA Grace CPU with a NVIDIA Blackwell Ultra GPU.
- 252 GB HBM3E GPU memory, delivering up to 7.1 TB/s memory bandwidth for demanding AI workloads.
- 496 GB ECC LPDDR5X CPU memory, providing up to 396 GB/s memory bandwidth.
- Up to 748 GB of coherent memory, enabling large AI models and datasets to be processed locally without relying entirely on cloud infrastructure.
- NVIDIA NVLink-C2C, providing high-speed communication between the Grace CPU and Blackwell Ultra GPU.
- Closed-loop liquid cooling, designed to support sustained high-performance AI workloads in a compact desktop form factor.
- High-speed networking, including two 400 Gb/s QSFP connections powered by NVIDIA ConnectX-8 SuperNIC technology.
- Flexible expansion, with PCIe Gen5 expansion slots and multiple high-speed M.2 storage interfaces.
- Optional NVIDIA RTX PRO GPUs, providing additional graphics and AI acceleration where required.
- 1600 W 80 PLUS Platinum power supply, designed for the system's high-performance computing requirements.
- Desktop AI supercomputer form factor, bringing Blackwell-class accelerated computing to local AI development, inference, research and enterprise environments.
- Designed for local AI, supporting large-scale model development, inference, generative AI, scientific computing and AI application development without requiring a full data-centre deployment.