Buy, Hold, or Sell: Is CoreWeave a Multibagger in the Making, or an AI Infrastructure Debt Trap?
CoreWeave (NASDAQ:CRWV) at $124.82 screens as a wait-and-see setup over the next 12 months, with a working reference value near $138.56. The stock has whipsawed since its 2025 IPO, with tension between a $99.4 billion backlog and $50.81 billion in…
CoreWeave (NASDAQ:CRWV) at $124.82 screens as a wait-and-see setup over the next 12 months, with a working reference value near $138.56. The stock has whipsawed since its 2025 IPO, with tension between a $99.4 billion backlog and $50.81 billion in liabilities driving the volatility.
CoreWeave rents GPU-accelerated cloud capacity to AI model builders and enterprises. Its customer roster includes Meta (NASDAQ:META | META Price Prediction), OpenAI, Anthropic, Mistral, Cohere, Jane Street, and Hudson River Trading, and it just crossed 1 GW of active power with 3.5 GW contracted and a target of 8 GW by 2030.
Why the multibagger case still has legs
Revenue grew 111.7% year-over-year to $2.08 billion in Q1 2026, beating consensus by 5.80%. Backlog jumped to $99.4 billion including a $21 billion Meta commitment. Management guided full-year 2026 revenue to $12 billion to $13 billion and 2026 exit ARR to $18 billion to $19 billion.
NVIDIA (NASDAQ:NVDA)’s $2 billion equity investment and CoreWeave’s designation as NVIDIA Exemplar Cloud for inference on the GB200 NVL72 validate the platform as the market shifts from training to inference. CEO Michael Intrator told analysts “we are now materially in excess of 50% of our compute being used for inference”, the stickier, higher-margin workload.
Why the debt math is loud
Total debt sits at $25.37 billion with capital lease obligations of $10.29 billion on equity of just $4.76 billion. Interest expense doubled year-over-year to $536 million in a single quarter, and the current ratio is 0.31x.
Free cash flow ran at negative $4.71 billion, and 2026 CapEx is guided at $31 billion to $35 billion. EPS came in at -$1.40, missing by 16.26%, the third miss in four quarters. If hyperscaler demand cools, every basis point on that debt stack matters.
Why patience beats conviction here
CFO Nitin Agrawal said “Q1 was the trough of our margin story”, with sequential expansion expected as deployments season. Q2 revenue is guided to $2.45 billion to $2.6 billion. Catalysts to watch: gross margin recovery in Q3, OpenAI capacity ramp, and whether the new investment-grade DDTL 4.0 facility priced below 6% proves the funding model can scale without further dilution.
A clean Q3 with rising contribution margins and on-schedule backlog conversion would strengthen the bullish thesis. A margin miss, hyperscaler cancellation, or refinancing scare would tilt the risk/reward the other way. Neither outcome is obvious yet.
What the consensus actually says
CRWV trades at $124.82 against an average analyst target of $138.56, implying roughly 11% upside. Of 34 covering analysts, 22 rate it Buy or Strong Buy, 10 Hold, and 2 Sell or Strong Sell. Targets are one data point among many.
Valuation reads at 9.6x sales and 30x EV/EBITDA, with no meaningful P/E given losses. Year-to-date, CRWV is up 74.31%, while SPY has gained only 2.8% since the May earnings report. The stock still sits 8.8% below its filing-day close of $136.80.
At $124.82, CoreWeave is a wait-and-see setup
The bull case and the bear case are both real, and the next two quarters will decide which one wins. A 12-month working reference near $138.56 captures the modest re-rating likely if Q3 confirms margin inflection and the DDTL playbook keeps compressing the cost of debt.
The key signals that would strengthen the bullish thesis are a Q3 print showing contribution margins back in the mid-20s, another step down in weighted average cost of debt, and backlog conversion tracking the guided 36% recognition over 24 months. The signals that would weaken it are any hyperscaler trimming capacity, interest expense outpacing adjusted EBITDA growth, or refinancing friction on the $25.37 billion debt stack.
The cost of patience is missing a melt-up if Anthropic, OpenAI, and Meta accelerate spending faster than guided. The cost of acting too early is owning a name where retained earnings are -$3.38 billion and current liabilities dwarf current assets if AI capex cools. Until Q3 settles the margin question, the setup respects both the opportunity and the leverage. Investors should consult a financial adviser before making any decision.
Contact [email protected] for any questions or corrections.
Alex Sirois is a financial writer with experience spanning both retail and institutional investing. He has written for InvestorPlace and held roles at BNY Mellon and Bernstein, giving him a perspective that bridges Main Street portfolios and Wall Street analysis.
Alex holds an MBA from George Washington University and has built his career across multiple industries, including e-commerce, education, and translation — a breadth of experience that informs how he breaks down complex financial topics for everyday investors. His writing is conversational, actionable, and grounded in long-term, buy-and-hold investing principles.
At 247 Wall St., Alex focuses on delivering analysis that is both accessible and useful, with a clear emphasis on helping readers make more informed decisions with their money.
