more
See Options
Need help? Ask our experts.
Review Summary
Accidents happen where life happens. Protect your favorite Adorama products and purchase a protection plan to stay covered from accidental damage, mechanical or electrical failure, and more.
If we can’t fix it, we’ll replace it at no additional cost.
Protect your product from drops, spills, and more.
Enjoy your gear without fear. We have you covered!
When regular use of your product over time results in mechanical or electrical failure.
We will never charge you a deductible after the purchase of a plan.
Our claims process is simple & easy and our customer service team is happy to help.
Adorama Protect powered by Extend is available for purchase to customers in the United States. Not available for purchase Internationally or in U.S. Territories.
Browse our FAQ
NVIDIA GB10
20-core Arm, 10 Cortex-X925 + 10 Cortex-A725 Arm
No
NVIDIA
128 GB
LPDDR5x
Onboard Only/Not Upgradeable
No
4TB
1x 4 TB / M.2 NVMe SSD
1x M.2 Slot (PCIe 5.0) [Occupied]
No
No
4x USB-C 3.2 Gen 2x2 / Supports Video Alt Mode and Power Delivery
1x HDMI 2.1a Output
No
1x RJ45 (10 GbE)
2x QSFP56 (200 GbE)
No
No
No
Wi-Fi 7 (802.11be)
Bluetooth 5.4
No
No
Black
NVIDIA DGX
Kensington Lock Slot
240 W with USB-C
110 to 240 VAC, 50 / 60 Hz at 2.5 to 3.5 A
32 to 95 Degree F (0 to 35 Degree C) at 10 to 90% Humidity
BSMI, CE, FCC Class B, UKCA, VCCI
Taiwan
5.95 x 5.95 x 2.05" (15.11 x 15.11 x 5.21cm)
2.65 lbs (1.2kg)
824142447444
Unleash Petaflop Power at Your Desk
The MSI EdgeXpert delivers 1 petaFLOP desktop AI performance through the NVIDIA GB10 Grace Blackwell Superchip-the same power driving NVIDIA DGX Spark. With 128GB memory and pre-installed AI software stack, run cutting-edge models up to 200 billion parameters from DeepSeek, Meta, and Google locally and instantly. Enterprise-grade performance in a compact form that scales seamlessly from desktop to data center. The future of AI development isn't in the cloud-it's on your desk.
Unrivaled Scalability
200B Parameter Local Processing
Run massive LLMs locally for complete data security, ultra-low latency, and cost control across your entire AI workflow.
405B Parameter Stacking
NVIDIA ConnectX networking enables seamless dual-system configuration to handle even larger AI models locally.
Seamless Scaling
Easily scale from desktop to NVIDIA DGX cloud, or any accelerated data center or cloud infrastructure.
Unmatched AI Computing Power
NVIDIA GB10 Grace Blackwell Superchip
Grace Blackwell architecture accelerates model tuning and real-time inference.
128GB LPDDR5X Unified Memory
Large memory capacity enables high-efficiency inference.
NVLink-C2C Technology
High-performance CPU to GPU interconnect with up to 5x PCIe 5.0 bandwidth.
1 petaFLOP of AI Tensor Performance
Blazing-fast FP4 processing effortlessly handles complex AI workloads at scale.
Edge Applications
Develop edge applications with NVIDIA AI frameworks & scale seamlessly to data center or cloud
MSI EdgeXpert provides an excellent platform for developing robotics, smart cities, and computer vision solutions. NVIDIA frameworks, including Isaac, Metropolis, and Holoscan enable developers to take advantage of the power of NVIDIA DGX architecture to quickly develop edge applications locally, and scale seamlessly to DGX cloud, or any accelerated data center or cloud infrastructure.
https://us.msi.com/page/warranty
888-447-6564