MSI XpertStation WS300 on NVIDIA DGX Station™ Architecture
สอบถามสต็อกสินค้า สอบถามเพิ่มเติม https://m.me/gearlabth หรือแอดไลน์ @gearlab
A next-generation deskside AI supercomputer built on NVIDIA DGX Station GB300 architecture, delivering data-center-class AI performance in a compact tower. Purpose-built for large language models (LLMs), generative AI, data science, and autonomous AI agents — bringing the full power of a data center to your desk, without cloud dependency.
Key Highlights
- NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip
- Up to 748 GB coherent unified memory
- Up to 20 PFLOPS AI performance
- NVIDIA ConnectX-8 SuperNIC — dual 400GbE, up to 800 Gbps networking
- Liquid-cooled CPU, GPU, and networking
- NVIDIA DGX OS + full NVIDIA AI software stack (NemoClaw / OpenShell)
Specifications
- Superchip: NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip
- CPU: NVIDIA Grace — 72-core Arm Neoverse V2
- GPU: NVIDIA Blackwell Ultra (single)
- Combined TDP: up to 1,400W
- Memory: 748 GB coherent unified memory (496 GB LPDDR5X for Grace CPU @ 396 GB/s + 252 GB HBM3e for GPU @ 7.1 TB/s) via NVLink-C2C
- Networking: 2× 400G QSFP112 ports (ConnectX-8 SuperNIC, 800 Gbps), 1× 10GbE RJ45 (Marvell AQC113)
- Storage: Up to 4× M.2 NVMe (PCIe Gen5 & Gen6)
- Cooling: Integrated liquid cooling (CPU, GPU, ConnectX-8, optical cages)
- Management: BMC with mini-DisplayPort output
- Power Supply: 1600W 80 PLUS Titanium ATX (requires 240V)
- OS: NVIDIA DGX OS
- Dimensions: 245 mm (W) × 528.4 mm (H) × 595 mm (D)
- Model / Item: XpertStation WS300 / SYSMSIWS301R
Ideal For
AI research labs, ML engineering teams, and data-sovereignty-focused organizations running LLM training and fine-tuning, generative AI, data science, real-time inference, and physical AI / robotics — deployed securely on-premises.
All specifications subject to change without notice. Please contact Gearlab for current pricing, configuration, and availability.