NBIS
Nebius Group N.V.
AI FOUNDRY / NEOCLOUD · INFRASTRUCTURE
~$206
NASDAQ · ~$51.9B mkt cap 1
52W: $32.88 – $214.44
Next earnings: Q2 2026 (~Aug 2026, date TBD)
👁 WATCH
Entry suspended — CapEx + execution risk reassessed
STATUS UPDATE — 14 May 2026
Verdict moved from Add (pullback) → Watch. Q1 2026 fundamentals validated the thesis; what changed is the action. The CapEx step-up to $20–25B (from $16–20B) compounds execution complexity across an already-aggressive 3-continent buildout — Missouri, New Jersey, Finland, Iceland and now the new 1.2 GW Pennsylvania site stacked on top. At the prior entry zone of $145–175, a single major site slip drops the bear case from −64% to −70%+; the risk asymmetry has shifted the wrong way. Suspend entry at any price until Q2 print (~Aug) + Missouri energisation cadence give the execution flag a clean read.
Q1 2026 validated the thesis decisively: the market continues pricing NBIS as a momentum neocloud, while the correct frame is a capital-compounding AI utility with $46B+ in contracted backlog (Meta $12B committed + $15B option, Microsoft $19.4B), NVIDIA Exemplar Cloud status on GB300, and a $9.3B cash fortress matched against pre-sold 2027 capacity. The variant perception is not growth — it is the structural premium of balance-sheet architecture, owned European data-sovereignty infrastructure, and a hidden NAV floor (~$6B+ in ClickHouse ~28% and Avride ~83% stakes) that screeners continue to ignore even after ClickHouse's $15B Series C valuation.2, 7
Fwd P/S (2026E) ⚠
~16×
EV/S ~16× · $3.31B 2026E consensus 3
Rev YoY (Q1 2026)
+684%
QoQ +75% ↑ · $399M vs $51M Q1 2025 2
Gross Margin
68.63%
Q1 2026 · ↓ from 70% Q4 2025 2
Cash / Runway ⚠
$9.3B
CapEx $20–25B 2026E · match-funded vs backlog 2
EPS Revision (90d)
↑ improving
Beat −$0.23 vs −$0.78 est · 12 analysts 3
Next Earnings
~Aug 2026
Q2 FY2026 · Date TBD
AI Rev % of Total
98%
$390M Nebius AI / $399M group Q1 2
CapEx 2026E
$20–25B
↑ from $16–20B · demand-driven raise 2
Thesis Durability
B+
Variant perception (unchanged): the market prices NBIS as a momentum neocloud; the correct frame is a capital-compounding AI utility. Q1 2026 materially strengthened this: ARR at $1.9B (end Q1) is already ahead of the $1.6B thesis-breaker threshold, and the revised Meta contract structure ($12B committed + $15B capacity option at Nebius' discretion) provides more contractual permanence than previously understood.
TAM: outsourced AI compute growing toward $200B+ by 2028 — but the durability of the neocloud window remains conditional on hyperscalers not absorbing 70–80% of that capacity by 2028–2029.
Moat upgrades post-Q1: NVIDIA Exemplar Cloud status on GB300 for training (the only non-hyperscaler designation); >3.5 GW secured with >75% from owned/contracted capacity (not leased); the 1.2 GW Pennsylvania owned site structurally separates Nebius from lease-dependent peers.
Bear stress-test: hyperscaler in-sourcing compresses neocloud TAM 70–80% by 2028–2029. Current price implies a 5+ year premium franchise. The Q1 beat extends the runway but does not resolve the 2028+ expiry risk.
TAM: outsourced AI compute growing toward $200B+ by 2028 — but the durability of the neocloud window remains conditional on hyperscalers not absorbing 70–80% of that capacity by 2028–2029.
Moat upgrades post-Q1: NVIDIA Exemplar Cloud status on GB300 for training (the only non-hyperscaler designation); >3.5 GW secured with >75% from owned/contracted capacity (not leased); the 1.2 GW Pennsylvania owned site structurally separates Nebius from lease-dependent peers.
Bear stress-test: hyperscaler in-sourcing compresses neocloud TAM 70–80% by 2028–2029. Current price implies a 5+ year premium franchise. The Q1 beat extends the runway but does not resolve the 2028+ expiry risk.
Thesis validated by Q1 execution. Window risk remains real but structural moat is building faster than expected.
Business Quality
B+
Q1 2026 beat: revenue $399M (+684% YoY, +75% QoQ); beat consensus ~$379–389M. Nebius AI adj. EBITDA margin 45% — ahead of the 40% full-year guide after just one quarter. Group adj. EBITDA $130M (vs $15M Q4, vs −$54M Q1 2025). Net income $621M, but includes $780.6M non-cash ClickHouse revaluation gain — strip this and the operating loss persists.2
Balance sheet step-change: cash $9.3B (from ~$3.7B at YE2025), bolstered by $4.3B convertible notes closed March 2026 at 1.25%/2.60% coupons — structurally cheap debt matched against pre-sold backlog. ClickHouse stake re-valued at ~$4.2B (28% × $15B valuation after Jan 2026 Series C).2, 7
Execution flags: gross margin ticked down 68.6% vs 70% Q4 — early sign of mix shift or cost inflation; CapEx raised to $20–25B (from $16–20B), increasing execution complexity on an already-aggressive 3-continent buildout.
Balance sheet step-change: cash $9.3B (from ~$3.7B at YE2025), bolstered by $4.3B convertible notes closed March 2026 at 1.25%/2.60% coupons — structurally cheap debt matched against pre-sold backlog. ClickHouse stake re-valued at ~$4.2B (28% × $15B valuation after Jan 2026 Series C).2, 7
Execution flags: gross margin ticked down 68.6% vs 70% Q4 — early sign of mix shift or cost inflation; CapEx raised to $20–25B (from $16–20B), increasing execution complexity on an already-aggressive 3-continent buildout.
Flow / governance notes:
- No material insider open-market buying or selling in last 90 days from open-access data ⚠.
- $4.3B convertible notes (Mar 2026) + $2B NVIDIA prefunded warrants — dilution path is live; manageable if revenue ramp sustains.
- ClickHouse and Avride valuations are company estimates / third-party marks — not independently verified from primary sources ⚠.
- Yandex lineage: ESG/compliance exclusion risk persists — structurally caps institutional ownership below 60–70% norms.
Margins are running ahead of guidance; cash fortress has been built. Non-cash income inflates Q1 headline; gross margin compression bears watching in Q2.
Entry Price Discipline
C
Multiples: Fwd P/S ~16× on $3.31B 2026E consensus. EV/S ~16× (EV = $51.4B after $9.3B cash vs $8.4B debt). Premium vs CoreWeave (~1.9× P/S) remains extreme — partially justified by cleaner balance sheet, contracted backlog, and NVIDIA partnership; not fully justified at 8× premium.
vs revised DCF: At ~$206, the stock sits above the revised bull case (~$150–180/share). Bull revision was driven by: (1) higher cash position raises NAV floor by ~$8–10/share; (2) ClickHouse stake revaluation adds ~$6/share to NAV. These raise the floor but not the ceiling — operating business DCF remains unchanged at 15% discount rate and 5% terminal growth.
2027 P/S compression path: consensus 2027E revenue ~$10B. At current price, 2027 fwd P/S compresses to ~5× — defensible for an infrastructure compounder. The question is whether the consensus holds.
vs revised DCF: At ~$206, the stock sits above the revised bull case (~$150–180/share). Bull revision was driven by: (1) higher cash position raises NAV floor by ~$8–10/share; (2) ClickHouse stake revaluation adds ~$6/share to NAV. These raise the floor but not the ceiling — operating business DCF remains unchanged at 15% discount rate and 5% terminal growth.
2027 P/S compression path: consensus 2027E revenue ~$10B. At current price, 2027 fwd P/S compresses to ~5× — defensible for an infrastructure compounder. The question is whether the consensus holds.
Flow signals (post-Q1):
- +15.8% on earnings day (13 May); intraday ATH $214.44. Price slightly retreated to ~$206 close.
- BofA raised PT to $205 (from $175) pre-earnings; 12 analysts consensus PT $170.29 — now below trading price; upgrades likely imminent.4
- BlackRock +9.43M shares (38,000%+ position increase) per 31 Dec '25 13F-HR 45-day lag.
- Beta ~3.1; elevated borrow rates; high retail/WSB interest — volatility is structural, not temporary.
Watch — no entry at any price for now. Even at the revised bull DCF zone ($150–180), $20–25B CapEx layered onto a 3-continent buildout (now incl. 1.2 GW Pennsylvania) shifts the risk asymmetry: a single major site slip drops the bear case toward −70%+. Suspend entry pending Q2 execution evidence.
Verdict — Watch, no entry at any price for now
The analysis is unchanged; the action is. Q1 2026 delivered a clean beat across all tracked thresholds — revenue $399M (vs $280M thesis-breaker floor), ARR $1.9B (vs $1.6B thesis-breaker), Nebius AI adj. EBITDA margin 45% (ahead of the 40% FY guide). The capital story materially improved: $9.3B cash, ClickHouse stake at $4.2B, $27B Meta contract with explicit option structure. Thesis durability is intact.
What changed is risk-asymmetry, not fair value. The CapEx guide raised to $20–25B (from $16–20B) compounds execution complexity across a 3-continent buildout that just added a 1.2 GW Pennsylvania site to a queue already containing Missouri, New Jersey, Finland and Iceland. Each incremental gigawatt is a step-function in execution risk. At the prior $145–175 entry, the bear case was −64%; with the new CapEx perimeter the bear case extends to −70%+ on a single major site slip. That is the wrong asymmetry to underwrite at any price right now — the binding constraint is execution evidence, not price.
Stay out for now. Re-evaluate after Q2 print (~Aug) confirms (a) Missouri energisation on-track, (b) ARR exit rate ≥$2.5B, (c) no further unfunded CapEx step-up. If all three clear, reconsider entry — but at a discount to the prior $145–175 zone, not at it.
What changed is risk-asymmetry, not fair value. The CapEx guide raised to $20–25B (from $16–20B) compounds execution complexity across a 3-continent buildout that just added a 1.2 GW Pennsylvania site to a queue already containing Missouri, New Jersey, Finland and Iceland. Each incremental gigawatt is a step-function in execution risk. At the prior $145–175 entry, the bear case was −64%; with the new CapEx perimeter the bear case extends to −70%+ on a single major site slip. That is the wrong asymmetry to underwrite at any price right now — the binding constraint is execution evidence, not price.
Stay out for now. Re-evaluate after Q2 print (~Aug) confirms (a) Missouri energisation on-track, (b) ARR exit rate ≥$2.5B, (c) no further unfunded CapEx step-up. If all three clear, reconsider entry — but at a discount to the prior $145–175 zone, not at it.
Size to 0% NLV. No entry on pullback. Re-evaluate after Q2 2026 print + Missouri energisation confirmation. CapEx + execution complexity are the binding constraints, not price discipline.
Opportunity-cost check
▾
vs CSPX (40% NLV)
NBIS remains the cleanest pure-play neocloud on contracted backlog, NVIDIA preferred-partner status, and owned European sovereign infrastructure. But with the $20–25B CapEx perimeter and 5-site buildout, the conditional downside (bear case −70%+ on a single site slip) outweighs the conditional upside (bull case 0–15% above entry) relative to CSPX's diversified compounding. Watch status until execution evidence clears.
vs current holdings
No direct overlap with TSM/AVGO/MU/META/GOOGL/NOW. Closest factor: TSM and AVGO (AI compute/infra layer). At Watch, NBIS is a deferred-decision name, not an alternative capital sink — the dry powder freed by the PLTR exit is better deployed into existing high-conviction holdings (META in-zone, NOW staged at sub-$90) than initiated into NBIS pre-execution-evidence.
Thesis-breakers
▾
Three falsifiers. Any one trips → do not enter. Any one tripping post-entry → exit. Updated after Q1 2026 print; Q1 threshold (01 below) was cleared at $399M / $1.9B ARR.
- 01Q2 2026 revenue <$550M (implies QoQ deceleration) OR ARR exit rate <$2.5B at end of Q2 — growth trajectory is bending, not accelerating toward $7–9B YE target · By Q2 2026 print (~Aug 2026)
- 02Adj. EBITDA margin tracking <25% for Nebius AI in Q2 or Q3 (vs 45% Q1 and 40% FY26 guide) — Q1 margin beat was structural, not a one-off; a reversal signals cost overruns or pricing pressure · By Q3 2026 print
- 03Additional non-match-funded equity raise >$2B (ATM draws or new convert not backed by specific customer contract) OR CapEx guidance raised again above $25B without corresponding ARR guide increase — dilution overhang materialises without revenue cover · By FY26 year-end
Key risks
▾
- R1 Execution on $20–25B CapEx plan (raised from $16–20B). Nebius must connect 800 MW–1 GW by end of 2026 across a now-larger 3-continent footprint, with the new 1.2 GW Pennsylvania site added to the queue. Each incremental gigawatt adds execution complexity. A major site delay misses the $7–9B ARR target; at current valuation a 30% ARR miss implies >40% downside based on comparable infrastructure growth-to-multiple re-rates.
- R2 Customer concentration — Meta + Microsoft = ~92% of backlog. Meta contract is now clarified as $12B committed + $15B option at Nebius' discretion — the option structure provides flexibility but also means Meta could exercise less than the full $27B. The committed $12B is for NVIDIA Vera Rubin infrastructure starting early 2027; any delay in Vera Rubin GPU availability cascades directly into Nebius revenue recognition.
- R3 Dilution path is active. $4.3B convertible notes (Mar 2026) + $2B NVIDIA prefunded warrants + live ATM program. At ~$52B market cap, each additional 5% dilution = ~$2.6B value transfer. The converts are structurally match-funded, reducing risk vs typical pre-profit dilution, but the aggregate dilution pool is large.
- R4 Yandex geopolitical overhang — ESG/compliance exclusion risk. CEO Volozh's Russian citizenship renunciation and OFAC clean-break status have not eliminated this: some LP mandates and ESG-screened institutions structurally cannot hold NBIS. This caps institutional ownership below 60–70% norms and creates periodic forced-selling episodes — a real multiple limiter that will not resolve quickly.
- R5 Hyperscaler in-sourcing — the 2028 window. AWS, Azure, Google and Meta are all expanding GPU capacity. By 2028–2029, their own Blackwell and post-Blackwell deployments may absorb 70–80% of demand currently outsourced to neoclouds. At $206, the market is pricing a 5+ year premium franchise. The $27B Meta contract commitment is structurally protective but does not resolve the question of what happens at renewal (2031+).
DCF scenarios
▾
Discount rate: 15% — speculative / pre-profitability profile; high execution risk, capital intensity, Yandex governance overhang. Terminal growth: 5%. FCF modelled from adj. EBITDA minus maintenance CapEx from 2028, assuming CapEx normalises to ~25–30% of revenue as infrastructure matures. NAV floor added: ClickHouse ~28% stake (~$4.2B at $15B valuation) + Avride ~83% stake (~$2.0B est ⚠ indicative) net of debt/cash. Revised upward from prior bull $130–160 / bear $50–70 to reflect $9.3B cash position and ClickHouse revaluation; discount rate and operating assumptions unchanged.
| Assumption | Value | Note |
|---|---|---|
| 2026E Revenue | $3.4B | Top of guide; Q1 $399M annualises to ~$1.6B, management guiding 3× ramp H2 |
| 2027E Revenue | $10B | Consensus; $1.9B ARR × ramp trajectory |
| 2028–2030 CAGR | 35–40% | Deceleration from hyper-growth as capacity fills |
| Normalised FCF Margin (2028+) | 20% | Adj EBITDA 40–45% minus maintenance CapEx ~20–25% |
| Discount Rate | 15% | Pre-profitability speculative; high execution risk |
| Terminal Growth | 5% | Infrastructure compounder; compute is durable even if neocloud window narrows |
| NAV Floor (ClickHouse + Avride) | +~$25/share | Portfolio assets net of debt vs cash |
| Implied Fair Value | ~$150–180/share | Revised up from $130–160; driven by NAV floor improvement |
Current price (~$206) is above the revised bull-case range. The Q1 beat raised the floor (via cash position + ClickHouse NAV) but not the operating-business ceiling. The stock is pricing either a sub-10% discount rate or 7%+ terminal growth — still not defensible for a pre-profitability, capital-intensive neocloud.
| Assumption | Value | Note |
|---|---|---|
| 2026E Revenue | $2.5B | 25% miss vs guide midpoint; major site execution delay |
| 2027E Revenue | $5B | Neocloud window compression; Meta option partially exercised |
| 2028–2030 CAGR | 20% | Hyperscaler in-sourcing headwind accelerates |
| Normalised FCF Margin (2028+) | 10% | Margin shortfall; ongoing dilution; cost overruns |
| Discount Rate | 15% | Pre-profitability speculative |
| Terminal Growth | 4% | Commoditising compute; niche European positioning |
| NAV Floor (net) | +~$10/share | Lower on asset-value write-down risk; cash consumed by CapEx |
| Implied Fair Value | ~$55–75/share | Modestly revised up from $50–70; higher cash NAV partially offsets |
Bear case implies 64–73% downside from current levels. Only requires one of: Q2/Q3 ARR miss, major CapEx execution delay, or Meta capex deceleration trimming option exercise. Q1 beat reduces probability but does not eliminate it.
SUMMARY: Bull case revised to ~$150–180 (from $130–160); bear case ~$55–75 (from $50–70). Current price (~$206) remains above both scenarios. The NAV floor has materially improved post-Q1 ($9.3B cash + ClickHouse revaluation), but the stock is still pricing 2028+ franchise durability that is not yet proven. Entry at $145–175 offers the margin of safety the bull DCF requires.
Catalyst timeline
▾
-
13 May 2026 ✓Q1 2026 earnings — BEAT. Revenue $399M (vs ~$389M est); ARR $1.9B (vs $1.6B thesis-breaker); Nebius AI adj. EBITDA 45% (ahead of 40% FY guide). Meta contract structure confirmed ($12B committed + $15B option). CapEx raised to $20–25B on pre-committed 2027 demand. Thesis-breaker 01 cleared.
-
Q2/Q3 2026Missouri data centre energisation — on-track confirmation (Q3) is the first real test of 3-continent buildout execution. Slip to Q4 = first CapEx execution flag.
-
~Aug 2026Q2 2026 earnings — next major gate. Watch: (1) revenue ≥$550M + ARR ≥$2.5B confirms glide path to $7–9B YE target; (2) Nebius AI adj. EBITDA margin ≥35% sustains thesis. Miss on either → thesis-breaker 01 trips.
-
H2 2026 / Early 2027Meta Vera Rubin GPU delivery cadence — NVIDIA GB300 (Blackwell Ultra) deployment schedule for the $12B committed Meta capacity. Any pushout from "early 2027" to late 2027 cascades directly into NBIS revenue recognition and ARR trajectory.
Named analyst commentary
▾
"Nebius is the purest Neocloud and AI-infra asymmetry left. This is the highest revenue Neocloud… with no plaguing uncertainty that $IREN, $ORCL face from full-stack execution. No high interest debt that $CRWV, $APLD, and others face… $8B midpoint ARR next year, $4.7B+ in cash, diversification in enterprise clients from Meta, Microsoft, Accenture, Shopify, Governments, hyper-growth portfolio companies, and a proven full-stack high-margin business."
Serenity (@aleabitoreddit) · May 2026 · PT: $400 / 1Y from post · At time of post: ~$86–92
"$NBIS vs. $IREN. The difference is night and day. Nebius: $2B dilution from $NVDA, zero immediate selling pressure to the public float… Very clear which company leads to higher share value appreciation from capex financing. One is strategic with Nvidia, the other is toxic financing."
Serenity (@aleabitoreddit) · 11 March 2026 · Commenting on NVDA $2B strategic investment
"Incredibly undervalued and detached from fundamentals. $7–9B forward ARR, 20–30% EBIT, enterprise contracts from Shopify, Accenture, Cursor, foreign governments and hyperscaler contracts from Meta and Microsoft give Nebius revenue visibility. With $4.8B+ in cash, it's isolated from credit tightening affecting data centres."
Serenity (@aleabitoreddit) · November 2025 · Rating: Strong Buy
Beth Kindig (@bethkindig) — No public commentary on NBIS found on free-tier Substack/X as of May 2026. IO Fund focus is AI infrastructure bottlenecks; neocloud operators are outside their typical coverage lane.
Dylan Patel / SemiAnalysis — No coverage of NBIS found on newsletter.semianalysis.com free tier as of May 2026. Focus is semiconductor supply chain; neocloud operators are outside scope.
Sources
1 Yahoo Finance / TradingView — NBIS quote, market cap, 52W range (accessed 2026-05-14) live data
2 Nebius Group — 6-K Q1 2026 financial results, SEC EDGAR / BusinessWire, filed 2026-05-13
3 Stockanalysis.com / consensus aggregators — fwd P/S, 2026E revenue $3.31B, analyst count 12, consensus PT $170.29 (accessed 2026-05-14)
4 Bank of America Research / TheStreet — PT raise to $205 (from $175), 11 May 2026
5 BigGo Finance / Investing.com — Q1 2026 earnings call transcript summary, 13 May 2026
6 Serenity (@aleabitoreddit) — X posts, Nov 2025 – May 2026
7 StockTitan / SEC 6-K — ClickHouse $400M raise at $15B valuation (Jan 2026); Nebius ~28% stake implied value ~$4.2B ⚠ company-confirmed post-revaluation
8 Fintel.io / Whale Wisdom — 13F-HR filings for period ending 2025-12-31 45-day lag
⚠ = figure not independently verifiable from primary open-access sources. Net income $621M includes $780.6M non-cash ClickHouse revaluation gain — operating loss persists on a cash basis. Avride stake valuation (~$2.0B) is indicative and unverified from primary source.
1 Yahoo Finance / TradingView — NBIS quote, market cap, 52W range (accessed 2026-05-14) live data
2 Nebius Group — 6-K Q1 2026 financial results, SEC EDGAR / BusinessWire, filed 2026-05-13
3 Stockanalysis.com / consensus aggregators — fwd P/S, 2026E revenue $3.31B, analyst count 12, consensus PT $170.29 (accessed 2026-05-14)
4 Bank of America Research / TheStreet — PT raise to $205 (from $175), 11 May 2026
5 BigGo Finance / Investing.com — Q1 2026 earnings call transcript summary, 13 May 2026
6 Serenity (@aleabitoreddit) — X posts, Nov 2025 – May 2026
7 StockTitan / SEC 6-K — ClickHouse $400M raise at $15B valuation (Jan 2026); Nebius ~28% stake implied value ~$4.2B ⚠ company-confirmed post-revaluation
8 Fintel.io / Whale Wisdom — 13F-HR filings for period ending 2025-12-31 45-day lag
⚠ = figure not independently verifiable from primary open-access sources. Net income $621M includes $780.6M non-cash ClickHouse revaluation gain — operating loss persists on a cash basis. Avride stake valuation (~$2.0B) is indicative and unverified from primary source.