Nebius Stock Has Nearly Doubled This Year. Here's Why It Still Has Room to Run.

Nebius is projected to grow revenue from $530 million at the end of 2025 to $9.7 billion by the end of 2027, implying nearly 20x growth, with analysts expecting 522% revenue growth in 2026 and 195% in 2027. The stock remains expensive at 73x sales, but the article argues that valuation may look more reasonable on forward sales if the company can scale toward profitability. Management is expanding data centers from 7 in 2025 to 16 by end-2026, reinforcing the aggressive AI infrastructure buildout.

Analysis

NBIS is the kind of name where the market usually misprices the transition from scarcity to scale: the first leg is narrative-driven multiple expansion, but the second leg depends on whether capacity can be monetized fast enough before supply catches up. The real bull case is not just revenue growth; it is that AI infrastructure demand remains structurally short relative to available compute, allowing a young operator to lock in long-duration customer relationships before pricing normalizes. That creates a potential “optionality premium” if management can keep utilization high while expanding footprint faster than competitors.

The key second-order risk is that the next 12-18 months may look better on top-line than on economics. Heavy buildout can compress near-term returns, and investors may eventually penalize NBIS if incremental data-center additions dilute ROI before contracts mature. In that scenario, the stock becomes highly sensitive to any slowdown in bookings, higher capex intensity, or evidence that hyperscalers/large cloud players are absorbing the same demand at better unit economics.

On the competitive side, NVDA benefits indirectly from every incremental AI server deployment, but NBIS’s outperformance can also signal a broader capex cycle that supports the whole AI stack. INTC is more of a lagging beneficiary only if the buildout shifts toward alternative silicon or edge inference later; near term, the market is still rewarding pure exposure to constrained AI compute. The contrarian miss here is that the market may be extrapolating revenue growth without fully discounting dilution, financing needs, and the probability that a great growth curve does not automatically translate into a durable terminal margin structure.

Key Decisions for Investors

• Maintain a tactical long NBIS only on pullbacks, not chase strength: entry on a 10-15% drawdown from recent highs with a 6-12 month horizon. Upside remains meaningful if utilization stays tight, but position size should be capped because the stock is effectively a levered claim on execution.
• Pair trade: long NBIS / short a basket of slower-growth cloud or infrastructure names over 3-6 months only if you want pure AI infra beta. Use a small net exposure because the thesis is timing-sensitive and vulnerable to any capex disappointment.
• Consider a call spread on NBIS rather than outright equity for 6-9 months. This captures another leg of multiple expansion if growth stays above consensus, while limiting downside if the market starts pricing dilution or weaker unit economics.
• For relative value, keep NVDA as the cleaner expression of AI capex durability versus NBIS as the higher-volatility operating leverage trade. NBIS has higher upside from sentiment, but NVDA has better downside protection if the AI spending cycle stays intact.
• Do not add aggressively until the market gets visibility on post-buildout margins and financing. If management shows that new data-center capacity is reaching payback within a defined window, the valuation regime can re-rate further; if not, the stock can de-rate quickly despite strong revenue growth.