Axe Compute Stock Targets Growth With $3.45 Billion in AI Contracts

Axe Compute (NASDAQ:AGPU) is building its business around a key opportunity in artificial intelligence: supplying the computing infrastructure that enterprises need to develop and run AI applications. Its strategy combines long-term customer contracts, dedicated GPU clusters and customer prepayments to support expansion in a market where infrastructure requires substantial upfront investment.

The company reported more than $3.45 billion in signed AI infrastructure contracts over seven months as of September 2026. It also disclosed more than $184 million in customer prepayments and over $266 million in additional contracts since its second-quarter results.

These figures give Axe Compute a substantial commercial foundation for its next stage of growth. The opportunity is to turn those commitments into operational computing capacity, recognized revenue and cash flow as demand for AI infrastructure expands.

Axe Compute’s AI Contracts Drive Recurring Revenue Growth

Axe Compute’s AX Build business is central to its growth strategy. It involves designing, deploying, owning and operating dedicated GPU clusters for customers under long-term agreements, solving a major challenge for enterprises adopting AI.

Building dedicated computing infrastructure requires substantial capital, technical expertise and access to facilities with sufficient electricity and cooling. Specialist operators can take responsibility for the computing equipment, allowing customers to secure dedicated capacity without managing the entire infrastructure themselves.

CEO Chris Miglino said in an interview that the company had signed approximately $3.5 billion in contracts and expected about $750 million in annual recurring revenue at the beginning of 2027. In August, the company had reported that its earlier contracts were expected to support an annualized revenue run rate exceeding $696 million upon full deployment.

The figures reflect the company’s expanding contract base and expectations for future revenue, rather than revenue already earned. Contract value covers commitments over multiple years, while annual recurring revenue represents the expected yearly run rate once the relevant infrastructure is operational.

That distinction defines the opportunity ahead. As more clusters go live, Axe Compute can begin recognizing revenue under its Build agreements and increase the scale of its operating business.

Customer Prepayments Can Reduce the Cost of Expansion

Building GPU clusters requires significant capital before the equipment starts generating revenue. Operators must purchase processors, install networking systems and prepare infrastructure before customers can use the computing capacity.

Axe Compute’s customer prepayment model helps address this funding requirement. In July, the company announced that it expected more than $534 million in aggregate customer prepayments associated with major Build contracts, alongside project-level financing to fund a substantial portion of the related infrastructure spending.

The company subsequently reported receiving more than $184 million in customer prepayments in the period covered by its September update.

These payments can reduce the amount of capital the company must raise independently to deploy infrastructure. They also connect financing more directly to customer demand, giving the business a way to expand against identified commercial commitments instead of relying entirely on speculative capacity purchases.

The central advantage is that customer commitments can help support infrastructure investment before the resulting revenue is fully recognized.

GPU Ownership Could Strengthen Project Economics

Owning computing hardware gives infrastructure providers a direct stake in the assets generating revenue. When companies combine that ownership with long-term customer agreements, they can align equipment investment with expected payments over the contract period.

Axe Compute’s Georgia AI cluster illustrates how ownership and operating arrangements can shape project economics. In October, the company holding the cluster and repaid approximately $87.8 million in asset-backed financing.

It also disclosed a further $42.9 million deferred purchase price, payable in installments. The company expects the cluster to generate approximately $364.6 million in revenue over its five-year customer agreement.

The arrangement also provides for Duos to cover facility and energy costs for the contract term, giving Axe Compute greater visibility into two important operating expenses.

This structure matters because revenue alone does not determine a project’s financial value.

For AI infrastructure companies, combining hardware ownership with contracted demand and more predictable operating expenses can support better financial planning. For AI customers, partnerships between equipment owners and data center operators can provide access to dedicated computing capacity without requiring them to build and manage every part of the infrastructure themselves.

Hardware Upgrades Can Protect Long-Term Revenue

Hardware ownership alone will not determine success in AI infrastructure. The rapid development of AI chips also makes hardware upgrades an essential part of the infrastructure business.

As newer processors deliver greater computing power and efficiency, older GPUs risk losing their competitive edge before infrastructure providers have fully recovered their initial investment.

Companies that plan for these upgrades from the beginning can protect customer relationships and maintain the commercial value of their computing capacity.

This is something Axe Compute has foreseen and planned for. The company has contract provisions that allow customers to move to newer hardware for an additional fee.

With this, they gain access to improved performance without having to establish an entirely new infrastructure arrangement, while the infrastructure company also recovers part of the cost of replacing equipment through the additional fee.

The approach is important as AI workloads expand beyond model training into inference, fine-tuning and production applications. Businesses running these workloads need reliable computing capacity that can adapt to changing performance requirements.

Paid upgrade provisions can help preserve customer relationships and reduce the risk that older equipment becomes a constraint on future contracts.

By incorporating technological change into its commercial agreements, an infrastructure provider can plan equipment replacement as part of its operating model rather than treating it as an unexpected expense.

Final Thoughts

Right now, the opportunity is to turn this multibillion-dollar contract portfolio into a larger, revenue-generating infrastructure business. As more clusters become operational, the company can begin demonstrating the commercial value of its customer commitments, hardware ownership and financing model.

That transition could become the defining development in the company’s growth story, positioning it to capture more of the spending required to support AI adoption.