Why Did NeoCloud Experience the Largest Gain in This Round of Tech Stock Rebound in the U.S.?

In this round of tech stock rebound in the U.S., one of the strongest directions comes from NeoCloud: CoreWeave, Nebius, and some AI infrastructure companies with power and data center resources.

Logically, funds are pricing a multi-leveraged AI infrastructure equity certificate: computing power capacity that has already been contracted and can be delivered quickly.

Once AI demand is revised upwards, NeoCloud's revenue expectations, financing capabilities, and shareholder equity value may all rise simultaneously. This gives it strong upside elasticity during the tech stock rebound phase; power, data centers, financing, and valuation elasticity together form this layer of leverage.

The bottleneck of AI is changing. Initially, the most scarce resource was GPUs, followed by HBM and high-speed networks; now, what customers truly lack is a complete set of deployable capabilities: obtaining GPUs, having sufficient power, completing data center construction, interconnecting networks, and being able to deliver large-scale clusters within months.

NeoCloud is precisely positioned at this gap.

Funds are buying "powered computing factories"

The products offered by NeoCloud typically include GPU clusters, networks, liquid cooling, data centers, power access, and operation and maintenance services. Customers purchase a large-scale computing capacity that can directly run AI training and inference.

This is crucial. GPUs can be procured, but power capacity, land, substations, data center licenses, and network access cannot be replicated in the short term. Large cloud vendors have capital and customers but are also constrained by construction cycles; some AI companies wish to retain more flexibility and are reluctant to place all their demands on a single hyperscaler.

Thus, NeoCloud, with ready power and rapid deployment capabilities, has become the "accelerator" for AI infrastructure investment.

The market is willing to give it a higher valuation, primarily because these resources have two characteristics:

· Scarcity: Available power and deliverable data center capacity are limited;

· Contractual: Customers are willing to sign long-term capacity contracts with minimum commitments.

When scarce resources can be locked in by long-term contracts, the market will reinterpret them from ordinary IT service revenue to cash flow assets with infrastructure attributes.

Financial Reports Changed Market Perceptions of Business Models

Previously, the market's main question about NeoCloud was straightforward: does buying GPUs and building data centers require huge Capex, leading the company into a cycle of "continuous financing and continuous cash burn"?

Recent financial reports have provided a relatively positive answer.

CoreWeave's Q2 revenue reached $2.575 billion, disclosing a backlog (expected revenue that has been signed but not yet confirmed) of approximately $104 billion; Nebius's AI Cloud ARR (annual recurring revenue) reached $3 billion, revealing multiple large long-term contracts. The market focuses on quarterly revenue but is more concerned about the complete business loop behind these numbers:

AI customers sign long-term capacity contracts
→ Some customers provide prepayments or minimum payment commitments
→ The company finds it easier to obtain debt and equipment financing
→ New GPUs, data centers, and power capacity come online
→ Revenue and EBITDA (earnings before interest, taxes, depreciation, and amortization) grow
→ Financing capabilities and expansion capabilities continue to improve

This shifts NeoCloud's narrative from "high Capex GPU lessor" to "AI infrastructure operator supported by orders."

As long as orders, financing, and delivery can continue to connect, growth exhibits a clear flywheel characteristic.

Why Did Funds Not Prioritize Storage and the Three Major Clouds?

The choice of funds reflects the expected differences at different stages.

Storage leaders benefit from AI demand, and the prosperity of products like HBM and DRAM remains high. However, the market has begun to worry about supply ramp-up, high prices, peak profit margins, and whether previous optimistic expectations have been fully reflected in stock prices. Strong financial reports can lead to stock price pressure if forward guidance does not continue to be revised upwards.

The challenge for storage companies lies in their cyclical nature. The market trades on the price, shipment volume, and gross margin paths for the next few quarters; when supply may catch up with demand and average selling prices may decline, strong current performance is unlikely to sustain valuation expansion. HBM/DRAM, NAND/SSD, and HDD also belong to different sub-cycles, and one cannot attribute all storage companies' stock price performance to the same reason.

The three major clouds—Microsoft Azure, Amazon AWS, and Google Cloud—have more robust cash flows, customer bases, and technical capabilities, and are also core beneficiaries of AI investment. Their AI businesses are diluted by large revenue bases from advertising, enterprise software, e-commerce, and consumer businesses; new AI capital expenditures will take longer to reflect in the overall group's profit margin improvement. For funds seeking elasticity, a large contract from NeoCloud often has a marginal impact on revenue and valuation greater than an equivalent order's impact on the overall valuation of the three major clouds.

NeoCloud sits between the two: it has a lower revenue base, a pure AI exposure, rapid order growth, and each new long-term contract may directly support the next round of financing and expansion. Funds are likely to view it as a highly elastic AI infrastructure target.

The current market trading logic can be summarized as:

NeoCloud is essentially a form of AI infrastructure leverage

Understanding why NeoCloud leads the charge hinges on grasping its leverage effect. Buying stocks in such companies essentially means holding an equity asset that is highly sensitive to AI computing power demand, deliverable capacity prices, and financing environments. The leverage here encompasses three layers of meaning.

The first is operational leverage. The upfront investment in GPUs, data centers, power access, networks, and operations is high, and many costs remain relatively fixed after capacity goes online. When the utilization rate of enabled clusters rises and unit capacity prices improve, new revenue can convert to profit at a faster pace, and the marginal improvement in profit margins will be significant.

The second is financing leverage. Long-term contracts, non-negotiable commitments, and customer prepayments can enhance the project's attractiveness to lenders and equipment financiers. As a result, the company can leverage part of its equity capital to facilitate larger-scale GPU, data center, and power investments; once new capacity begins to bill, revenue can support the next round of construction.

The third is equity leverage. NeoCloud's revenue base and market value are typically smaller than those of the three major clouds, yet fixed assets and debt account for a higher proportion on the balance sheet. A large contract that simultaneously enhances revenue expectations, utilization rates, and financing availability can lead to a steep revaluation of shareholder equity value. The rapid rise in stock prices after financial reports often results from the combination of upward revisions in profit expectations and valuation multiples.

These three layers of leverage create positive feedback during the upward phase:

Larger long-term contracts
→ Easier access to financing and capacity expansion
→ Higher utilization and operational profits
→ Increased equity value and financing capabilities
→ More contracts and opportunities for the next round of expansion

The same mechanism can also amplify downside risks. If customers delay, utilization rates decline, GPU or power deliveries lag, or debt costs rise, fixed costs and financing obligations will compress shareholder returns. Therefore, the high elastic pricing given to NeoCloud by the market also reflects its high execution requirements.

Order visibility is the core of this round of revaluation

The most attractive part of NeoCloud lies in the visibility of its revenue.

If customers sign non-negotiable contracts, even if actual usage fluctuates in the short term, customers still need to bear a certain minimum payment obligation. For operators, this type of revenue is easier to predict; for creditors, these contracts also enhance the feasibility of asset financing.

Thus, the market will continue to track several indicators:

· The duration, binding force, and customer credit of signed contracts;

· The gap between enabled MW (megawatts) and contracted MW;

· Revenue per unit MW and Capex per unit MW;

· The proportion and payment rhythm of customer prepayments;

· Utilization rates, renewal rates, and customer concentration;

· Debt interest rates, debt terms, and subsequent financing capabilities.

Among these, "enabled capacity" is particularly critical. The contracted MW represents demand, while the MW that is powered, installed, and begins billing will enter revenue and cash flow.

This is also a revaluation of power assets

The most valuable insight about NeoCloud in the community is shifting the focus from the number of GPUs to Power (power capacity).

The supply of GPUs will expand with the procurement of NVIDIA, AMD, and cloud vendors; however, the formation of high-quality power capacity is slower. It involves power grids, substations, land, permits, data center construction, and regional network conditions.

Whoever can secure sufficient power earlier can convert GPUs into sellable computing power sooner.

This is also why some companies that have transformed from Bitcoin mining can enter this main line: they already have some power resources, land, and infrastructure, only needing to shift their assets from mining loads to AI loads. Of course, resource foundations do not equate to business success; ultimately, it still depends on customers, financing, and delivery capabilities.

NeoCloud's leading position in the tech stock rebound is backed by the market's reordering of the AI infrastructure value chain.

Currently, the most valued assets by funds are those that can combine GPUs, power, data centers, and long-term customer contracts and deliver computing capacity quickly. It not only undertakes AI capital expenditures but also possesses stronger contractual characteristics than mere chips and components; it has both high growth elasticity and a premium for infrastructure scarcity.

Moving forward, whether NeoCloud can continue to outperform depends on a very simple question: can these huge orders be transformed into powered clusters, confirmed revenue, and cash flow that covers capital costs on time?