Welcome to our dedicated page for SuperX AI Technology news (Ticker: SUPX), a resource for investors and traders seeking the latest updates and insights on SuperX AI Technology stock.
SuperX AI Technology Limited (NASDAQ: SUPX) is a Singapore-headquartered AI infrastructure solutions provider focused on full-stack systems for AI data centers. The news flow around SUPX centers on its transition from a legacy interior design business to a strategy built on high-performance AI servers, 800 Volts Direct Current (800VDC) power solutions, high-density liquid cooling, AI cloud, and AI agents.
News updates for SUPX frequently highlight product launches and technology milestones. These include announcements of next-generation AI servers, rack-scale platforms such as the SuperX GB300 NVL72 System powered by NVIDIA GB300 Grace Blackwell Ultra Superchips, and the introduction of Modular AI Factory solutions that integrate compute, cooling, and power. Coverage also features the launch of 800VDC power products like the SuperX Panama-800VDC and SuperX Aurora-800VDC solutions through the SuperX Digital Power joint venture, as well as liquid cooling initiatives via the SuperX Cooltech joint venture.
Investors following SUPX news will also see updates on strategic partnerships and ecosystem developments, such as the majority investment in MicroInference, a solution provider in the NVIDIA Partner Network, and the formation of SuperX Global Service with Teamsun to deliver global end-to-end AI infrastructure services. Corporate governance and leadership changes, including the appointment of Dr. Chenhong Huang as Chairman and CEO and the appointment of Ken Lau as Chief Strategy Officer and Executive Director, are disclosed through press releases and related SEC filings.
Capital markets activity, including private placements of ordinary shares and warrants and the authorization of a share repurchase program, appears in both press releases and Form 6-K filings. For users tracking SUPX, this news page provides a consolidated view of product announcements, joint ventures, leadership updates, financing transactions, and other corporate developments related to the company’s AI infrastructure strategy.
SuperX AI (NASDAQ: SUPX) announced that on November 19, 2025 MicroInference Pte. Ltd., a Singapore-based subsidiary majority owned by SuperX, was elevated by NVIDIA in the NVIDIA Partner Network Solution Provider Partner Program.
The elevation, effective November 19, 2025, applies to competencies including Compute and Networking at the Preferred level. As a Preferred level partner, MicroInference may be eligible for enhanced program benefits. A copy of the NVIDIA promotion letter is furnished as Exhibit 99.1 to the Form 6-K filed with the SEC.
SuperX AI Technology (Nasdaq: SUPX) appointed Dr. Chenhong Huang as Chairman, Executive Director, and CEO effective December 1, 2025. The company said Dr. Huang brings more than 30 years of multinational enterprise technology leadership across SAP, Dell, Schneider Electric, Tellabs, and Nortel, and will drive SuperX's global rollout of a modular AI Factory and full‑stack AI infrastructure strategy.
The appointment follows two recent board appointments on November 11, 2025 and is positioned to strengthen executive management, corporate governance, and deployment of SuperX's proprietary liquid cooling and energy‑efficient data center technologies.
SuperX (NASDAQ: SUPX) reported FY2025 results reflecting its legacy interior-design operations while announcing a strategic shift to full-stack AI infrastructure.
Key FY2025 figures: Revenue $3.6M (including ~$1M AI equipment sales in June 2025), Net loss $21.2M, Cash $17.2M, and Assets $52.1M. Post-year-end (Jul–Oct 2025) moves include joint ventures for HVDC power and liquid cooling, acquisition of a majority stake in MicroInference (NVIDIA partner), a Japan supply center and U.S. subsidiary planned by end-2025, new AI servers and rack-scale products, and investor commitments totaling >$240M.
SuperX (NASDAQ: SUPX) announced two 800VDC full-chain power products—Panama-800VDC (end-to-end for new builds) and Aurora-800VDC (modular retrofit)—aimed at resolving high-density GPU power bottlenecks on Oct 31, 2025. The systems claim up to 98.5% system efficiency, support single-rack power demands to 3.6 MW (Panama) and scale from 200 kW to 1 MW+ (Aurora), and reduce copper usage by over 45%. Panama uses a native DC path from medium-voltage input to GPU with modular, hot-swappable design and prefabricated deployment; Aurora enables non-disruptive single-rack upgrades within hours. Both solutions are compatible with NVIDIA Kyber rack architecture and target hyperscale AI and retrofit data centers.
SuperX (NASDAQ:SUPX) and Teamsun announced formation of a Singapore joint venture, SuperX Global Service, with SuperX AI Solution owning 51% equity. The JV will provide end-to-end AI infrastructure services—global contact centers, deployment, maintenance, and managed operations—for SuperX Modular AI Factories, AI servers and third-party AI products.
Teamsun brings scale: presence in 18 countries, >7,000 employees, 37 delivery centers, and service history with >16,000 enterprise clients. The JV aims to close SuperX's product+service loop, accelerate global deployment, and offer standardized turnkey service products to support SuperX's global strategy.
SuperX (NASDAQ: SUPX) announced a strategic investment giving it a majority shareholding in MicroInference, a Singapore-based NVIDIA Partner Network solution provider, effective Oct 24, 2025.
The move is presented as part of SuperX's long-term plan to build a high-performance AI ecosystem across Asia Pacific by strengthening its supply chain for NVIDIA servers and networking, accelerating deployment of Modular AI Factories, and gaining prioritized access to NVIDIA technical expertise, certifications, and support.
SuperX (NASDAQ: SUPX) and a Hong Kong subsidiary of Chengtian Weiye announced on October 24, 2025 the formation of a Singapore joint venture, SuperX Cooltech, with SuperX as the single largest shareholder. The JV will develop and supply rack-level liquid cooling products and infrastructure globally, excluding mainland China, Hong Kong SAR, and Macau SAR.
Key capabilities include Micro Channel Liquid Cold Plates (MCLP), Cooling Distribution Units (CDUs), manifolds and liquid-cooled racks, combined with system integration and mass-production capacity to offer end-to-end liquid cooling solutions, shorten delivery cycles, and improve data-center energy efficiency.
SuperX (NASDAQ: SUPX) launched the SuperX GB300 NVL72 System on October 16, 2025, a liquid-cooled, rack-scale AI platform built around the NVIDIA GB300 Grace Blackwell Ultra Superchip.
The NVL72 rack links 72 Blackwell Ultra GPUs and 36 Grace CPUs to deliver up to ≈1.8 exaFLOPS FP4 AI compute, ≈165TB HBM3E and ≈17TB LPDDR5X memory per rack, 800Gb/s InfiniBand XDR networking, and a 48U MGX rack form factor. SuperX positions the system for hyperscale, sovereign AI, exascale scientific computing and industrial digital twins, and bundles the rack with liquid cooling and 800VDC power infrastructure as part of a prefabricated modular AI factory solution.
SuperX (NASDAQ: SUPX) on October 3, 2025 launched the XN9160-B300 AI Server, an 8U flagship built around the NVIDIA Blackwell B300 (HGX B300) targeting large-scale AI training, inference and HPC.
Key specs: 8 Blackwell B300 GPUs (288GB HBM3E each; 2,304GB unified HBM3E), Dual Intel Xeon 6 CPUs, up to 32 DDR5 DIMMs, 8×800Gb/s OSFP networking, NVLink, eight Gen5 NVMe bays, and 12×3000W 80 PLUS Titanium redundant PSUs. NVIDIA cites +50% NVFP4 compute and +50% HBM per chip vs prior Blackwell. The server is positioned for hyperscalers, scientific research, finance, bioinformatics, and global systems modeling.
SuperX (NASDAQ:SUPX) has unveiled its groundbreaking SuperX Modular AI Factory, a revolutionary data center-scale solution designed to transform AI infrastructure deployment. The solution integrates compute, cooling, and power systems into pre-fabricated modules, reducing traditional data center deployment time from 18-24 months to under 6 months.
The system features impressive specifications including up to 20MW per module, supporting up to 144 NVIDIA GB200 NVL72 systems, and achieving a power efficiency (PUE) as low as 1.15, resulting in 23% energy savings compared to traditional systems. The solution's modular design allows for flexible 1-to-N scaling, while its high-density compute capabilities deliver 7 times higher rack power density than traditional data centers.