All CategoriesBuilt for How Business Buys.
Deskside AI Supercomputer
ASUS ExpertCenter Pro ET900N G3
Data Center-Class AI, at Your Desk
NVIDIA® GB300 Grace Blackwell Ultra Desktop Superchip
Up to 748 GB Large Coherent Memory
Up to 20 PFLOPS of Al performance
NVIDIA Grace 72-Core Neoverse V2 Processor
NVIDIA
ConnectX-8 SuperNIC
Up to 800Gb/s for multi-system scale-out
Trillion-Parameter AI, Developed Locally
Built from the ground up to build and run AI, the ET900N G3 unleashes the NVIDIA® GB300 Grace Blackwell Ultra Desktop Superchip and 748GB of large coherent memory - enough to prototype, fine-tune and infer models of up to 1 trillion parameters at the deskside.
NVIDIA AI Enterprise and NVIDIA NemoClaw
Secure, production-ready agentic Al out of the box
NVIDIA® ConnectX®-8 SuperNIC
Links two systems to handle even larger models
Ubuntu with NVIDIA Al Developer Tools
Preconfigured stack, ready on first boot
Grace 72-Core Neoverse V2 processor · up to 4 x M.2 2280 NVMe PCle 5.0 SSD, 8TB total
Local Deployment, Enterprise Control
Keep regulated and proprietary data inside your own four walls. The ET90ON G3 runs training and inference on-premises, with predictable cost and full control over where your data lives.
Scales From One System to Many
Start at a single deskside system and grow into a cluster. QSFP 400G networking and the NVIDIA ConnectX-8 SuperNIC move data between systems fast enough to train on models no single machine could hold.
Extreme Cooling for Extreme AI Workloads
A data center-grade thermal design pulls cool air across the superchip and exhausts it clear of the chassis, so sustained AI workloads hold their clocks.
Managed Like the Rest of Your Fleet
ASUS Control Center Express
Inventory, monitoring, deployment and patching from one console
Ubuntu with NVIDIA AI Developer Tools
Preconfigured and supportable on day one
Dedicated BMC
Out-of-band management over 1G LAN, Micro-B or MiniDP
ASUS Commercial Support
Business-class warranty and service network
Agentic AI, Built In-House
Prototype, fine-tune and run fleets of AI agents on your own hardware - without sending a single token off-site.
NVIDIA NemoClaw and NVIDIA AI Enterprise
Secure, production-ready agentic AI, preconfigured
748 GB of coherent memory
Holds whole models and long agent context
Ports and Slots
Front
1Power button
2Message LEDs
3Audio
4USB 10G Type-C
5USB 10G Type-A
6USB 2.0
Back
1QSFP 400G ports
2USB 10G Type-A
31G LAN for BMC
410G LAN
5Audio
6Micro-B port for BMC
7MiniDP for BMC
Dimensions
