USSTOCK.TODAY
Regular Market
Log in Sign up
Is this stock a buy?

NBIS (Nebius Group): Should You Sell This Stock? — 5th Week of June (Issue 16)

$NBISLarge Cap
Telegram KakaoTalk
Briefs · earnings · signals, first Telegram · KakaoTalk channels

Buy side vs Sell side

🤔
Sell side
It doesn't matter how much revenue grows if nothing is left over. They sell 100 won worth of product and only pocket 7.5 won, while actually running a sizable operating loss. What they're scaling up is not growth—it's a money-losing business.
🤭
Buy side
Amazon Cloud also went through a phase of pouring money out at the start. This is just the stage of investing early in GPUs and data centers, so the loss is natural—once utilization ramps, profits follow. Cash on hand is plenty, so there's enough stamina to ride it out.
🤔
Sell side
That analogy doesn't hold. Amazon sold its own software from day one, so margins were high, but this company is in the rental business, leasing out somebody else's GPUs. On top of that, those machines age out in 3~4 years and lose value, so "margins will improve" is not a plan—it's wishful thinking.
🤭
Buy side
The NVIDIA CEO himself publicly promised "Nebius will take care of you." In an industry where the hardest thing to get is NVIDIA chips, having that supply secured is itself a powerful weapon. They're also preparing lucrative new businesses like healthcare.
🤔
Sell side
NVIDIA says nice things to anyone willing to buy chips—in fact, it's in NVIDIA's interest to nurture multiple competitors. Big players like SoftBank are flooding money into the same business, and Meta has even sent a signal of pausing capex for a moment. The healthcare business is still just a trailer with no revenue and no customers.
🤭
Buy side
Institutional ownership is over 50% and keeps climbing, and brokerages are issuing price targets far above current levels. Last quarter's results also beat expectations by a wide margin. That means smart money is already flowing in.
🤔
Sell side
You have to look at the other side too. Short positioning betting on the decline is over 25% of the float. That means a significant share of professional investors are confident "this stock can't hold up" and are positioning for it. The current picture—where only retail investors are packed with optimism—is more unsettling.
🤭
Buy side
That very short interest is the opportunity. If next quarter's results come out even slightly better, short sellers could scramble to cover and the stock could spike. The company is confident revenue will grow 500% this year, so there's a real chance of that.
🤔
Sell side
The visible profit is also an illusion. This year's net income is a number inflated by one-time gains, not by core operations. Strip that out, and the story is: buy a company with losses and heavy debt at over 60x revenue, while the growth rate professionals see for next year is only about 10%.
🤭
Buy side
You have to look at the substance of the debt too. It's all long-term debt, and it's debt used to buy GPU equipment that generates revenue. The company holds $37 per share in cash, so it's not a company that will wobble tomorrow. AI infrastructure demand is just getting started—this is buying cheaply with a 3-year view.
🤔
Sell side
"Believe in demand for the next few years and buy" is exactly what people investing in telecom networks during the 2000 dot-com bubble said. In a market where supply is flooding in, the first to collapse is exactly places like this one with thin margins. This stock right now isn't a discount item—it's a growth story attached to a falling knife.
🤭
Buy side
For a "falling knife," the company's stamina is far too strong. Cash is ample, supply chain is secured, and growth is showing up in actual numbers. Right now, when everyone is spooked, is precisely the moment to scoop up good companies cheaply. Having good stamina and having a profitable business are two different problems. Until the core business's money-making structure is proven, watching the next quarter to confirm margin improvement with your own eyes and then stepping in is not too late.
// i18n-ok: 한국어 본문을 파싱·치환하는 값 (화면 문구 아님)

30-Second Summary

The 오미주 AI Committee issued a Sell verdict on NBIS. Within the general-purpose GPU rental business structure operating at a 7.5% gross margin, the AWS-style J-curve logic does not hold; headline EPS is inflated by one-time gains, and the stock trades at roughly 62x real revenue with negative free cash flow. Compounding competitive and demand cracks—from SoftBank's 10GW entry to Meta's capex digestion—the recommendation is to immediately sell 1/3 at market, then ladder out the remainder over 5~7 trading days and keep only up to ~1% for tracking purposes.

Trading Guide

Item Detail
Action Sell
Entry — (Sell recommendation: 1/3 immediately at market + remaining 2/3 sold in tranches at higher limit prices on any short-squeeze rebound)
Stop-loss No explicit price — rather than a simple stop below the prior decline low, recommended is to protect the remainder by buying small put positions (price/ATR data not verified; history of −17% single-day gaps)
Target
Hold period Complete exit within 5~7 trading days; reassess at Q2 earnings (~2~3 months out)
Weight Residual exposure capped at ~1% NAV (only a small tracking position kept)

Bull Scenario

Revenue is on an explosive trajectory of +443.9% YoY, with the +500% revenue guidance and Jensen Huang's public endorsement ("Nebius will take care of you") backing the growth narrative. With cash per share of $36.95 and a fortress-like current ratio of 8.33, solvency risk is effectively ruled out, and institutional ownership is also trending up at 50.3%. The unveiling of a healthcare/life-sciences-focused AI platform is a signal of diversification beyond general-purpose GPU rental into a high-margin, high-stickiness vertical, and the Wall Street consensus is also maintaining buy ratings with a price target of $257.79.

Bear Scenario

The core issue is unit economics. With a gross margin of 7.5% and an operating margin of -68.8%—in a structure renting out general-purpose GPUs over a 3~4-year depreciation cycle—the mechanism is fundamentally different from AWS, which ran gross profit at scale from the early build-out phase. Headline EPS (~$3.00) and a 95% net margin are figures inflated by one-time non-operating gains; in substance, the market is only modeling forward EPS growth of +10.8% for a company trading at roughly 62x real revenue, with negative free cash flow and a debt-to-equity ratio of 1.31—and that gap is what this latest drop is really about. On top of that, SoftBank's 10GW neo-cloud commitment, NVIDIA's incentive to nurture many competitors, and Meta's capex-digestion phase are building pressure from both the competitive and demand sides.

오미주 AI Consensus

All three risk analysts converged on Sell, and the debate was only over execution—not direction. The portfolio manager cited the past SNDK lesson—"when headline earnings are not sustainable, it's not a weight reduction but a full sell"—and overrode the research manager's weight-cut view, upgrading it to Sell. Execution is the compromise of selling 1/3 immediately at market to confirm risk, then treating the 25.2%-short-interest short-squeeze rebound not as a precondition but as a bonus, laddering out the remaining 2/3 over 5~7 trading days. The residual ~1% is to be protected with puts, and any additional bearish bets should be expressed only via put spreads rather than naked shorts, due to squeeze risk.

Key Data at a Glance

Metric Value Implication
Market Cap / EV $54.3B / $54.5B AI infrastructure large-cap
P/E / P/S / EV/Sales 71.9 / 61.8 / 62.0 Extreme overvaluation pricing in flawless execution
Gross Margin / Operating Margin / Net Margin 7.5% / -68.8% / 95.3% Core operations loss-making; net income from one-time non-operating gains
ROE / ROIC 14.5% / 5% Capital efficiency still low
TTM Revenue / YoY $877.9M / +443.9% Hyper-growth but low-margin
Debt-to-Equity / Current Ratio 1.31 / 8.33 Debt-funded expansion; liquidity solid
Cash per Share / Book Value per Share $36.95 / $28.52 Strong cash cushion; low solvency risk
Free Cash Flow (FCF) Negative / not aggregable Heavy capex phase
Short Interest Ratio 25.2% (2.9-day cover) High squeeze/volatility risk
Consensus PT / Recommendation $257.79 / 1.94 (Buy) Wall Street still friendly—view gap with committee

⏰ Triggers & Checkpoints

Conditions That Would Break the Bull Scenario

  • Q2 results show gross margin stuck below 15% (key gate)
  • Close below the prior decline low
  • After Meta, additional capex-cuts/digestion signals from other hyperscalers
  • Visible GPU rental price competition from new neo-cloud supply such as SoftBank's 10GW
  • Healthcare platform delayed in preview state without any paying customer

Conditions That Would Break the Bear Scenario

  • Q2 results show gross margin reversing to 15~20%+ and utilization disclosed (re-entry condition 1)
  • Healthcare platform lands named customers with revenue (re-entry condition 2)
  • +500% revenue guidance achieved alongside confirmed margin improvement
  • A concrete path to FCF breakeven emerges
  • The 25.2% short interest drives a strong rebound exceeding the upper end of the sell ladder, sustained (requires re-tuning of exit pace)

Go Deeper

Let's unpack this ticker once more (optional)

💼 Company Size

Nebius Group (NBIS) is a Netherlands-headquartered AI cloud infrastructure company listed on Nasdaq with a market capitalization of approximately $54.3 billion, qualifying it as a large-cap. A relatively small organization of {{EMPLOYEES}} employees runs its hyperscale AI infrastructure business, with trailing twelve-month revenue of about $877.9 million.

The growth trajectory is nothing short of explosive. Revenue grew +443.9% year-over-year and +621.5% quarter-over-quarter, and EPS is also up +428.8% YoY. However, the headline net margin of 95.3% should be viewed as a number substantially inflated by one-time items such as asset sales and revaluations. The operating margin is -68.8%—a deep loss—and gross margin is just 7.5%, meaning the core business itself has not yet reached profitability.

Valuation is at quite stretched levels. P/E of 71.9x, P/S of 61.8x, EV/Sales of 62.0x, and an EV/EBITDA of an abnormal 9,390x—the latter meaning EBITDA is effectively near zero. Given that next year's estimated EPS growth is set at only +10.8%, the current share price is pricing in a scenario in which AI infrastructure build-out is executed nearly perfectly.

Financial stamina shows two sides. With a current ratio of 8.33 and cash per share of $36.95, the liquidity cushion is ample, but funding GPU and data-center investment with debt has pushed the debt-to-equity ratio up to 1.31. There is no dividend, and free cash flow is negative—classic heavy-capex phase. Institutional ownership stands at a rising 50.3% (+15.3%), while short interest reaches 25.2% of the float, making this a stock with high volatility and short-squeeze risk.

Category Metric Value Interpretation
Size Market Cap / EV $54.3B / $54.5B Large-cap AI infra
Growth Sales YoY / QoQ +443.9% / +621.5% Explosive hyper-growth
Growth EPS YoY +428.8% (EPS $3.00) Inflated by one-offs
Profitability Gross / Op / Net Margin 7.5% / -68.8% / 95.3% Core ops unprofitable; net income non-operating
Valuation P/E / P/S / EV/Sales 71.9 / 61.8 / 62.0 Extremely rich
Valuation EV/EBITDA 9,390x EBITDA near zero
Balance Sheet D/E / Current Ratio 1.31 / 8.33 Leveraged but liquid
Liquidity Cash/sh / BV/sh $36.95 / $28.52 Strong cash cushion
Ownership Insider / Institutional 19.7% / 50.3% (+15.3%) Institutions accumulating
Risk Short Float 25.2% High squeeze/volatility risk
Sentiment Analyst Rec / Target 1.94 (Buy) / $257.79 Street constructive
Cash Flow FCF Negative/NA Heavy capex phase

📈 Charts & Trends

In this analysis, there's something we need to be candid about. Repeated attempts to retrieve price data failed (the 2026-01-05~07-04 window, 5 attempts), so we could not secure a verified OHLCV time series or computed indicator values. Issuing a buy/sell direction with unverified data would be irresponsible, so our formal technical view is 'HOLD (pending).'

The indicator setup originally planned is informative as reference: 50-day and 200-day simple moving averages for medium- to long-term trend and golden/death cross context; 10-day EMA for short-term momentum turns; MACD, MACD histogram, and RSI for momentum direction and divergences; ATR for volatility-based stop width; and VWMA to confirm volume-backed trends.

As NBIS is a high-beta AI infrastructure name, the guiding principle is caution until data confirms the stock sits above its 50/200-day moving averages, the MACD histogram is positive, and the VWMA is supportive. Existing holders should keep the position with a stop set around ~2x ATR, while new entries should reasonably be deferred until the indicators confirm.

Item Detail
Ticker NBIS
Analysis window 2026-01-05 → 2026-07-04
Data status Retrieval failed (tool error, 5 attempts)
Selected indicators close_50_sma, close_200_sma, close_10_ema, macd, macdh, rsi, atr, vwma
Trend read Unconfirmed — pending data
Risk guidance ATR-based stops (~2x ATR); no new exposure until data confirms trend
Recommendation HOLD (pending indicator confirmation)
Next step Re-run get_stock_data + get_indicators with working data access

🎬 Pre-Announcement Mood

The news flow over the past week (6/27~7/4) is constructive. The core catalyst is the preview of the 'Nebius Scientific AI & Healthcare Platform'—a strategic move to diversify beyond plain GPU rental into a high-margin, high-stickiness vertical in healthcare and life sciences. On July 2, BNP Paribas published a new report covering CoreWeave and Nebius, showing that institutional coverage of the AI neo-cloud theme remains alive.

Retail sentiment is 'bullish with mixed undercurrents.' By directed-message count on Stocktwitz, bulls 11 vs. bears 2, so buying sentiment leads—but this is dip-buying character following roughly a 25% plunge over three weeks. Jensen Huang's remark that "Nebius will take care of you" and the +500% growth guidance are repeatedly cited as the bulls' basis, while anxieties over competitive intensification—such as SoftBank's AI cloud entry (10GW by 2030) and Meta's capex-cuts—are by no means minor talking points.

The caveat is that this bullish sentiment rests on thin foundations. There are only two news items, and over the past 7 days there have been zero NBIS mention posts on Reddit, so the signal relies on a sample of 30 Stocktwitz messages. An 85/15 bull skew immediately after a sharp drop can be an overheat signal, and as a cash-burn growth stock sensitive to rates, the name is also vulnerable to macro headwinds.

Catalysts:

  • Healthcare/Scientific AI platform launch (vertical expansion)
  • Potential new large compute contracts (expectation of "$300+ run" on new contract announcement)
  • Halo effect from relationship with NVIDIA
  • Momentum re-flow into AI infrastructure names after the drop

Risks:

  • Competitive intensification from SoftBank's 2030 10GW AI cloud push
  • Signs of hyperscaler capex cuts (Meta)
  • −25% drawdown over 3 weeks underscoring fragile flows
  • Direction of BNP Paribas report unconfirmed
  • 85/15 bull skew immediately after the drop = possible retail overheat
  • Absence of cross-verification from communities like Reddit

🧭 Conclusion in One Line

Explosive growth and institutional accumulation are attractive, but core-business losses, extreme valuation, and a 25% short interest are pressing back—so new buying is on hold and we recommend retaining HOLD on existing positions only. Consider scaling up only after gross-margin improvement and operating leverage are confirmed.


⚠️ Disclaimer — This article is decision-making content simulated by a multi-agent AI system and does not constitute investment advice. Entry, stop-loss, and target figures in the body are simulated numbers. All investment decisions and responsibility rest entirely with the investor.

Disclaimer: This content is for informational purposes only and does not constitute investment advice. All investment responsibility lies with the investor.

Today's 5 AI picks, all free
Nothing hidden: past picks and how they did against the S&P 500.
See today's picks →