Flex (NASDAQ: FLEX) announced on October 12, 2026, an expanded portfolio of prefabricated modular data center solutions designed to accelerate deployment of next-generation AI factories.

The centerpiece of the expansion is a new line of AI Modules that integrate power, high-density compute, and advanced liquid cooling into pre-engineered, factory-built systems.

According to the company, the approach enables customers to reduce deployment timelines by up to 50% compared with traditional construction approaches.

AI Infrastructure Expansion Portfolio Ahead of Axiom Spin-Off

The announcement, issued from San Jose, California, positions the expanded portfolio within Flex's Cloud and Power Infrastructure (CPI) segment.

Notably, Flex plans to separate that segment into an independent publicly traded company, Axiom Solutions International, Inc. (Axiom), in the first quarter of calendar 2027, meaning the new product line is closely tied to the forthcoming spin-off.

"Meeting the demands of AI infrastructure at scale requires more than incremental improvements. It requires rethinking how power, computing, and cooling work together as a system," said Chris Butler, expected Chief Technology and Strategy Officer at Axiom.

"Our system-level approach brings these critical technologies together with advanced manufacturing to simplify integration, accelerate deployment, and deliver AI factories at scale."

What the AI Modules include

The Axiom AI Modules integrate critical AI infrastructure into pre-engineered, factory-built systems spanning three core components.

First, an 800 VDC Power Rack delivers high-efficiency power conversion and distribution for next-generation 800 VDC AI infrastructure and high-density compute.

Second, advanced cooling distribution units (CDUs) provide liquid cooling to efficiently manage the increasing heat generated by power-intensive AI workloads.

Third, high-density IT racks integrate AI servers, networking, and high-speed interconnects to support next-generation AI compute at scale.

Factory-built systems address growing deployment complexity

Flex framed the launch against a backdrop of rising complexity for data center operators. As AI infrastructure grows in scale and power density, operators face increasing challenges across power delivery, cooling, compute integration, and deployment.

The company argues that meeting these challenges requires a more integrated approach to designing and deploying the infrastructure that supports AI factories. The key operational distinction in the new offering is a shift of complex infrastructure integration from the data center site to a controlled factory environment.

The Axiom AI Modules are assembled, integrated, and tested in parallel with site development. Flex says this parallel approach reduces on-site complexity and helps customers bring AI capacity online faster.

Manufacturing scale behind the push

Flex states it is actively building prefabricated modular solutions for hyperscalers across its global footprint, including more than three million square feet of dedicated data center infrastructure manufacturing capacity in the United States.

The company cites deep expertise across power, compute, cooling, and systems integration as the foundation for supporting rapid AI infrastructure deployment across regions.

The launch also strengthens Flex's broader Cloud and Power Infrastructure portfolio, which the company has committed to separating into Axiom in the first quarter of calendar year 2027.

Spin-off context and forward-looking caveats

The announcement arrives against the planned separation of the CPI business, a move Flex describes in forward-looking statements as intended to deliver benefits including enhanced strategic focus, financial flexibility, and value creation for shareholders.

The company cautions that the spin-off remains subject to a number of risks and uncertainties, including whether the spin-off will be completed and the timing thereof, and the possibility that various conditions to completion may not be satisfied or waived.

Additionally, there is the possibility that the spin-off will not qualify for the expected tax-free treatment for U.S. federal income tax purposes, and the risk that the separation may be more difficult, time-consuming, or costly than expected, including impacts on Flex's resources, systems, procedures, and controls.

Flex also warns that forward-looking statements about the anticipated benefits of the spin-off, the expected future performance of each company following completion, management changes and leadership, business strategies, growth opportunities, market position, and financial outlook are based on current expectations, estimates, and assumptions involving risks and uncertainties that could cause actual outcomes to differ materially from those anticipated.

Media inquiries for the announcement are directed to Flex's press contact, with investor and analyst relations handled by Michelle Simmons, Senior Vice President of Global Investor Relations and Public Relations.

Monitor All Hyperscale Developments Before Construction Begins Worldwide

Hyperscalers, colocation operators, and contractors are racing to deliver capacity across the globe, but the winners are the ones who spot opportunities first. Where will the next 100 MW campus rise, and which contractors have already secured the packages?

The Global Project Tracking (GPT) platform by Blackridge Research gives you a live, structured view of data center development across the world, from early-stage announcements to final commissioning, so your teams can move on leads months ahead of the market.

Every record is enriched with project timelines, owner and developer details, contractor assignments, and capital expenditure estimates, turning scattered announcements into a pipeline you can act on.

• Upcoming Projects

• Tender Notices

• Contract Awards

• Projects Under Construction

• Completed Projects

Book a Free Demo of the Global Project Tracking (GPT) platform by Blackridge Research and see the global data center pipeline before your competitors do.