Nebius Stock Surged 200% in Just 1 Year. Is It Still a Buy?

Nebius surged after Q2 results: revenue +454% YoY to $582M and adjusted EBITDA of $236M, alongside four AI cloud contracts worth $1B+ each. The company projects 2026 revenue of $3.0B–$3.4B and $7B–$9B of annual recurring revenue, with contract-related infrastructure payback improving to ~1 year 10 months vs ~2–3 years historically. However, the stock appears fully priced (about 18x forward P/S on 2026 top-end revenue) and it fell more than 20% last week following a $5B private convertible notes offering, raising dilution/debt concerns—leading the article to be bullish on fundamentals but cautious on valuation and near-term stock risk.

Analysis

NBIS is transitioning from a pure scarcity story to a capital-markets story: the market now has to underwrite not just demand, but the cost of financing that demand. The near-term winners remain upstream GPU and networking suppliers with pricing power, but the incremental loser is any AI infrastructure operator whose growth requires repeated external funding; as soon as customers/prepayments slow, equity becomes the shock absorber. That makes NBIS less a simple “AI winner” and more a levered duration bet on continued tight supply.

The main catalyst path is not the next quarter of revenue, but whether the company can keep turning backlog into contracted, prepaid, and high-ROIC capacity without another dilution event. Over the next 1-3 months, the key risk is multiple compression if investors conclude the stock is valuing perfect execution while ignoring capex intensity and convert overhang. Over 6-18 months, the structural bull case only holds if industry supply growth does not outrun demand; if hyperscalers and other neo-clouds keep adding capacity, pricing power can fade quickly and the payback math deteriorates.

Consensus is probably missing how reflexive this setup is: strong demand can actually invite the next wave of supply, which hurts margins before it shows up in headline growth. That said, the bear case is not about demand disappearing; it is about return on capital normalizing faster than revenue, which would compress the multiple even if the top line stays hot. In that sense, NBIS looks more expensive than fragile—fine for momentum, dangerous for new money unless the next financing and utilization data confirm the one-year-payback narrative.

Key Decisions for Investors

• Avoid chasing NBIS after the post-earnings rerating; wait for either a 15-20% pullback or the next financing update to confirm dilution is contained. Falsify the cautionary view if management reiterates contract-backed capacity with no incremental equity and payback stays under ~2 years.
• Pair trade: long NVDA / short NBIS for 1-3 months. NVDA monetizes AI capex with less balance-sheet risk, while NBIS still carries execution and financing risk; this works best if AI spending remains strong but the market punishes capital intensity.
• If already long NBIS, reduce into strength and replace part of the exposure with a basket of AI infrastructure beneficiaries that have stronger pricing power and less capex drag. The risk/reward on fresh NBIS entry is poor unless revenue guidance is revised materially higher.
• Set a hard alert on the next convertible or equity financing announcement: any deal that implies another meaningful dilution cycle should be treated as a short trigger, especially if the stock fails to hold recent support after the announcement.