ASUS used Computex 2026 to unveil Ascent QN10, billed as the world’s first AI mini PC built around Qualcomm’s Snapdragon X2 Elite with an 80-TOPS neural processing unit, plus a broader ‘enterprise-to-edge’ AI lineup that runs from server racks down to handheld devices. The June 2 announcement is part of ASUS’s pitch that the next AI infrastructure layer is being built at the edge as much as in the data centre.
Ascent QN10: An 80-TOPS Pocket AI PC
The QN10 is a compact desktop that pairs Snapdragon X2 Elite’s 80-TOPS NPU with low-power LPDDR5X memory, on-device Copilot+ support and an enterprise-targeted security stack. ASUS argues the form factor — a true mini PC for AI inference rather than a cut-down laptop board — is what changes the deployment story for edge AI: branches, retail back rooms, manufacturing floors and clinical workstations can host a local model without a data-centre round-trip.
The Enterprise-To-Edge Stack
ASUS framed the announcement under its Ubiquitous AI – Incredible Possibilities strategy, with the Computex line-up spanning AI servers built on NVIDIA Vera Rubin (announced this week), HBM-rich workstation towers, and the Ascent QN10 family. The company also showed updates to its Companion robot and Kairo service robot — both featured in our earlier coverage of ASUS’s Computex robotics debut — bringing the same edge-inference silicon into customer-facing platforms.
Why Snapdragon, And Why Now
Qualcomm has been signalling for two years that Snapdragon X-class chips would push from premium Windows laptops into mini PCs and edge appliances; the QN10 is the first marquee design landing in that segment. Combined with Snapdragon’s NPU-friendly tooling and the broader IE-IoT expansion Qualcomm completed earlier this year, ASUS’s mini PC is positioned as a serious challenger to Intel Core Ultra and AMD Ryzen AI Max boxes at the edge. For context: AMD’s newly launched Ryzen AI Max PRO 400 with 192GB memory targets local 300B-parameter models; the ASUS–Qualcomm QN10 targets always-on, always-quiet branch deployments.
Computex’s Edge Pivot
The QN10 reveal lands alongside Intel and Siemens’ industrial AI pact, Foxconn’s Intel infrastructure deal and Microsoft’s Project Solara chip-to-cloud OS. The pattern at this year’s show is a deliberate spread of AI workloads from hyperscale racks down to corridor-grade hardware — and ASUS is positioning the Ascent line to be that corridor.
Reporting based on coverage from ASUS Pressroom, AI Weekly and Computex 2026 keynotes.
