ASUS ExpertCenter Pro ET900N G3

ASUS ET900N G3 Extreme AI Performance: Powered by NVIDIA® GB300 Grace Blackwell Superchip delivering up to 20 petaFLOPS of AI performance and 748GB coherent memory for large AI development and inferencing workloads. • Developer-Optimized Platform: Tailored for AI researchers and developers with full-stack compatibility and streamlined workflows. • Scalable Architecture: Powered by NVIDIA® NVLink-C2C for ultra-fast CPU-GPU memory communication and NVIDIA® ConnectX-8 networking, enabling dual ASUS ExpertCenter Pro ET900N G3 system stacking to deliver greater memory capacity and model scalability. • Standard Tower, Advanced Performance: Enhanced compute power in a standard tower system, purpose-built for enterprise AI services, research, and large-scale model training. • Full Stack AI Solution: The Pro ET900N G3 NVIDIA AI Developer Tools provide a full stack solution for AI development and deployment Datasheet: [ASUS-ET900N-G3] The ASUS ExpertCenter Pro ET900N G3 is positioned as a deskside AI supercomputer rather than a conventional workstation, built for local AI training, inference, and agent workflows with up to 20 PFLOPS of AI performance and 748GB coherent memory. 6 Example Use Cases Private enterprise LLM deployment Run large language models locally for internal copilots, knowledge assistants, or document Q&A where low latency and data control matter. Autonomous AI agents for business workflows Build and operate secure agentic AI systems using NVIDIA NemoClaw/OpenShell for policy-controlled automation of research, ops, or support tasks. Computer vision for smart infrastructure Support local vision inference for monitoring, inspection, and infrastructure analytics, which aligns with ASUS’s smart infrastructure positioning and its listed computer vision support. Healthcare and medical research AI Use it for model development and inference in medical or research environments that benefit from on-prem execution and stronger control of sensitive data. Engineering and scientific computing Apply it to simulation-heavy or compute-intensive workflows in areas like aerospace and automotive, especially where AI and scientific computing overlap. Generative AI content creation Use it for local generative AI workloads such as image, media, or design-related content creation, which ASUS explicitly lists among supported AI applications.