ASUS announces enterprise agentic AI offering and Rubin servers
Reported by The Register and news.lavx.hu2 sources ↓
At NVIDIA’s GTC event in Berlin, ASUS announced a three-part enterprise AI offering and two HGX Rubin NVL8-based systems for AI reasoning workloads. The systems use different cooling designs, with one hybrid-cooled and one fully liquid-cooled.
Key points
- The offering combines ASUS AI Server and Infrastructure, AI Software Stack, and AI HUB services
- The XA NR11-E12L is a 3U hybrid-cooled system; the XA NR11-E12 is a 2U liquid-cooled system
- Both systems use an eight-GPU NVIDIA HGX Rubin NVL8 board
- ASUS says tool-less CPU and GPU sleds can cut maintenance time by up to 50%
Sources
Details
ASUS announced the Enterprise Agentic AI Solution at NVIDIA’s GTC event in Berlin on Oct. 21–22, according to both articles. It brings together server and infrastructure products, software for AI workflows, and services under ASUS AI HUB. The company says the offering combines existing products with new ones.
The two announced Rubin systems target large AI reasoning workloads. The 3U XA NR11-E12L uses direct-to-chip liquid cooling for its GPU baseboard and air cooling for its dual Xeon 6 chips; the 2U XA NR11-E12 uses full liquid cooling and a cable-free board-to-board design. Both articles report 800 Gbit of integrated bandwidth per GPU for the NVIDIA HGX Rubin board.
ASUS says the systems’ tool-less CPU and GPU sleds can reduce maintenance time by up to 50%. The articles do not give pricing or availability dates.
Written from coverage as of
Images
- The Register
- news.lavx.hu
Stocks in this story
Change since open: 20 Nov to 21 Nov 2026
Last refresh