NBIS institutional options flow analysis — multi-leg block trades, dominant direction, and gamma analysis from the public options tape for July 6, 2026. Articles older than 60 days are public; a free account reads back to 30 days, Pro to 5, and AIme Premium reads today's unusual options trades with no delay.

AInvestOption Pilot Program
Unusual FlowDaily DigestNBIS Ticker Page

NBIS Unusual Options Activity — 2026-07-06

Institutional flow on 2026-07-06

Multi-leg block trades, dominant direction, and gamma analysis

$18.6M1 trade
LEAP Call Block Cross (direction unprovable)

Trade Details

CROSS$340 CALL2028-12-15$18.6MLEAP Call Block Cross (direction unprovable)

Full Analysis

🌐 NBIS $18.6M LEAP Call Block Cross — A 2.4-Year Bet Nobody Can Prove Is Bullish (Yet) 🤝

📅 July 6, 2026 | 🔥 Unusual Activity Detected ✅ Updated 2026-07-07: next-day OPRA OI confirms this OPENED — the $340 LEAP-call strike's OI rose 60 → 1,510 (+1,450 ≈ the 1,500 block). Fresh open confirmed; direction still unprovable on a cross. See RESOLVED box below.

🎯 The Quick Take

A 1,500-contract block of NBIS December 15, 2028 $340 calls — a genuine LEAP, 2.4 years out — printed today at $124.00, exactly at the midpoint of the $122/$126 market. That's ≈$18.6 million changing hands in one negotiated trade. Here's the catch: this printed as a 🤝 block cross — a broker matched a known buyer and known seller off the lit book — which means there's no aggressor and no way to prove from the tape alone whether the buyer or the seller initiated it. This is a story about someone taking a multi-year view on Nebius's AI-cloud buildout, not a "smart money is loading up" headline. Let's walk through exactly what we can and can't say.

📊 Company Overview

Nebius Group N.V. (NBIS) is a vertically integrated "neocloud" — an AI and high-performance-computing infrastructure provider built around large-scale GPU clusters, cloud platform software, and developer tooling. The company is a carve-out of the former Yandex N.V.: in 2024 Yandex sold off its Russian assets and renamed the remaining international business Nebius Group.

  • Market Cap: ≈$56 billion (as of today's ≈$220.90 print; was ≈$66.3B at $261.15 on June 29 — the stock has pulled back roughly 15-20% in the past week)
  • Sector: AI-cloud / high-performance-compute infrastructure ("neocloud"), classified under software infrastructure
  • Current Price: ≈$220.90 at the time of this trade (intraday charts below were snapshotted a bit later at ≈$218)
  • YTD Performance: up +143% in 2026 (from ≈$89.95 to ≈$218.56), after peaking near $290 in late June
  • Core business: AI-cloud (GPU rental + managed inference), which made up 98% of group revenue in Q1 2026; also owns Avride (autonomous driving), TripleTen (edtech), and stakes in Toloka and ClickHouse

Nebius has spent the last ≈10 months stacking anchor AI-infrastructure contracts: Microsoft (up to $19.4B), Meta (up to $27B), and a $2B strategic equity investment from Nvidia. Q1 2026 revenue grew 684% year-over-year to $399M, and the core AI-cloud segment now runs a 45% adjusted-EBITDA margin.

💰 The Option Flow Breakdown

📊 What Just Happened

The Tape (July 6, 2026 @ 12:36:34):

TimeSymbolBuy/SellTypeExpirationPremiumStrikeVolumeOISizeSpotOption PriceOption Symbol
12:36:34NBIS🤝 CROSSCALL $3402028-12-15≈$18.6M$3401,500601,500$220.90$124.00NBIS20281215C340

🤝 Mechanism: single-leg block cross. This printed at exactly the midpoint of the $122.00 bid / $126.00 ask — the textbook signature of a pre-arranged, negotiated block where a broker paired a known buyer and known seller off the open book. There is no aggressor in a cross. The screenshot's "N/A" side label is correct, not missing data — this trade genuinely has no buy/sell tag on the tape. We are not calling this "aggressive buying." It isn't.

✅ RESOLVED — Fresh Open Confirmed (Direction Still Unprovable on a Cross)

The July 7 pre-market OPRA snapshot (reflecting July 6 end-of-day) is in. (July 3 was a full market holiday for the Independence Day observance, so the pre-trade baseline was July 2's end-of-day.)

LegBaseline (EOD Jul 2)Resolving (EOD Jul 6)ΔVerdict
Dec-15-2028 $340 Call601,510+1,450≈ the 1,500 block → OPENED ✅
  • Confirmed opened. OI jumped from 60 to 1,510 — a +1,450 rise, almost exactly the 1,500-contract block (the small shortfall of ≈50 reflects a bit of transfer against existing holders). New open interest was created; this was not a close.
  • But direction is still unprovable. This printed as a block cross with no aggressor. Confirming "open" does NOT resolve whether the buyer or the seller holds the bullish multi-year AI-cloud view — a cross has a known counterparty on both sides, and OPRA carries no side tag. That may never be provable from public data.

🤓 What This Actually Means — Plain English

Let's translate the jargon:

  • 🗓️ A "LEAP" is just a long-dated option — LEAP stands for Long-term Equity AnticiPation security, industry slang for any option with more than a year until expiration. This one has ≈2.4 years left (expires December 15, 2028). Compare that to the typical options trade you see in unusual-activity scans, which expire in days or weeks. Whoever is on the other side of this trade is planning around Nebius's business in 2028, not next week's headlines.

  • 🤝 A block cross has a buyer AND a seller who already agreed on price. Think of it less like "someone slammed the ask" and more like two institutions calling their broker and saying "let's do this trade privately, off the public order book, at a price we both already like." Because it printed at the exact midpoint ($124.00, dead center of $122/$126), neither side paid up or got a discount — a classic sign of a negotiated, not urgent, transaction.

  • ❓ Why we genuinely cannot tell you if this is bullish or bearish. For a lit sweep (someone hitting the offer on the public book), we can read the tape and say "that's a buyer taking liquidity." For a cross, that tool doesn't work — OPRA doesn't publish which side initiated a negotiated block. Two equally plausible stories fit the same print:

    • The bullish read: A fund or institution wants leveraged, multi-year exposure to Nebius's AI-cloud buildout without laying out the ≈$33M it would cost to buy 150,000 shares outright. For ≈$18.6M, they control the same share-equivalent upside above $340, with breakeven at expiration around $464 (strike $340 + premium $124) — well above even the current Street high target of $380.
    • The "premium seller" read: A long-time holder (an early investor, insider, or fund sitting on a huge unrealized gain after the +143% YTD run) is writing far-out-of-the-money calls — the strike is ≈54% above today's spot — to generate cash flow while keeping most of the near-term upside, essentially a long-dated covered-call overwrite.
    • Both stories are consistent with the same tape. We genuinely don't know which one is true, and we're not going to pretend otherwise.

  • 🧩 What OPRA cannot see, ever: the counterparty's identity, whether either side already owns NBIS stock or other options, whether this is part of a larger hedged package (a collar, a financing structure, a delta-hedge), or the broker/MMID involved. All of that stays private. What we can say with confidence: size (1,500) massively exceeds strike-level OI (60), so new contracts are almost certainly being created, and the multi-year AI-cloud backlog (Nvidia's $2B stake, Meta's up-to-$27B deal, Microsoft's up-to-$19.4B deal, with deliveries running into 2027) is exactly the kind of long-horizon story that would make someone want December 2028 exposure in the first place — from either side of the trade.

Unusual Score context: a 1,500-lot, $18.6M single print on a strike that had only 60 contracts of prior open interest is a genuinely large block for NBIS — the kind of size you see a handful of times a year on this name, not routine daily flow. Size alone tells you this was a deliberate, planned transaction; it doesn't tell you direction.

📈 Technical Setup / Chart Check-Up

YTD Performance Chart

NBIS YTD Performance

Nebius is up +143.0% year-to-date (from $89.95 to $218.56), but that headline number hides a wild ride: 105.4% annualized volatility and a -32.06% max drawdown along the way. The stock rocketed from the low-$100s in April to a peak near $290 in late June on the back of the Nvidia and Meta deal news, then pulled back roughly 20% to today's ≈$218-221 level over the past week. Today's LEAP cross prints right into that pullback, not at the top.

Gamma-Based Support & Resistance Analysis

NBIS Gamma S/R

Reference spot for this map: ≈$218.19 (snapshotted slightly after the trade, which printed with the stock at $220.90 — normal for data pulled at different points in the same afternoon).

🔵 Support Levels (Put Gamma Below Price):

  • $210 — nearest floor, ≈$0.89B in put gamma
  • $200 — ≈$2.01B put gamma, a meaningfully bigger step down
  • $190 / $185 — ≈$1.51B / ≈$0.98B, secondary shelf
  • $180 — the single biggest level on the entire chart, ≈$4.0B in put gamma. This is the line dealers will defend hardest if NBIS keeps sliding.
  • $170 — ≈$1.92B, a deeper floor beyond that

🟠 Resistance Levels (Call Gamma Above Price):

  • $220 — right at today's level, ≈$0.54B call gamma
  • $230 — ≈$0.53B
  • $240 — the strongest overhead call-gamma level, ≈$1.24B — the first real ceiling above spot
  • $250 / $260 — ≈$0.96B / ≈$0.60B, secondary resistance band
  • $270 / $280 / $300 — thinning out (≈$0.69B / ≈$0.64B / ≈$0.78B) but still notable extended targets

What this means for traders: Across the whole chain, total put gamma (≈$26.5B) roughly doubles total call gamma (≈$13.3B) — this is a put-heavy, hedging-dominated book, which tracks with a stock that just had a 20% pullback from its highs. The $180 level is the dominant gamma magnet on the downside; $240 is the first real overhead resistance on the upside. None of this tells us anything about the LEAP cross's direction — it's simply the shorter-dated options market's current view of near-term support/resistance, useful context for anyone trading NBIS this week, not this decade.

Implied Move Analysis

NBIS Implied Move

  • 📅 Weekly (Jul 10 — 4 days): ±$32.19 (±14.75%) → Range: $186.04 - $250.42
  • 📅 Monthly OPEX (Jul 17 — 11 days): ±$51.72 (±23.7%) → Range: $166.48 - $269.92
  • 📅 Quarterly Triple Witch (Sep 18 — 74 days): ±$123.92 (±56.79%) → Range: $94.28 - $342.12
  • 📅 1-Year LEAPS proxy (Jun 17, 2027 — 346 days): ±$251.95 (±115.47%) → Range: essentially $0 - $470.15

Translation for regular folks: Even at just the 1-year mark, the options market is pricing a swing wide enough to put NBIS anywhere from near-zero to north of $470 — that's how volatile this name is priced. The Dec 2028 LEAP in today's trade goes another year and a half beyond that already-enormous 1-year cone. This is the entire point of a LEAP: whoever holds either side of this $340 call is underwriting an extraordinary amount of uncertainty in exchange for extraordinary potential optionality, over a horizon where quarterly implied-move charts stop being useful and the real drivers become the multi-year AI-infrastructure buildout described in the catalysts below.

🎪 Catalysts

Important distinction: none of the dates below are the option's expiration. This LEAP expires December 15, 2028. The catalysts below are the fundamental events between now and then that will actually decide whether NBIS is worth more or less than $340 (plus the $124 premium, i.e. ≈$464 breakeven for a long call holder) by that date.

🔥 Confirmed Near-Term Catalyst

Q2 2026 Earnings — July 29, 2026 (confirmed)

Nebius reports 23 days from today. Watch for the core AI-cloud ARR trajectory versus Q1's $1.92B, progress toward the $7-9B annualized run-rate and $3.0-3.4B group-revenue guidance, capex pacing against the $20-25B 2026 plan, and an updated read on contracted power toward the >4 GW by end-2026 target. Some data aggregators list an alternate Aug 6 estimate — treat July 29 as confirmed and verify against Nebius Investor Events as the date approaches.

🚀 Already-Happened Strategic Developments (Last ≈10 Months)

📆 Expected Developments (H2 2026)

🗓️ Multi-Year Structural Drivers Through 2028 (the LEAP's actual lifespan)

  • Meta Vera Rubin deployment — deliveries start early 2027, squarely inside the LEAP's life.
  • Nvidia's >5 GW by 2030 roadmap brackets the December 2028 horizon.
  • Combined Microsoft + Meta backlog ≈$46B — signed capacity whose recognition over the next several years is the core thesis for anyone holding this LEAP long.
  • Analyst dispersion is wide — average price target ≈$244, with a range from a $120 low (Goldman Sachs, Sept 2025) to a $380 high, and recent marks of $287 (Citigroup, May 2026) and $280 (BofA, June 2026). Note that even the highest Street target ($380) sits below this LEAP's ≈$464 breakeven-if-long — reaching profitability on the bullish read requires the stock to outrun current analyst enthusiasm.

🎲 Price Targets & Probabilities (Through the 2028 LEAP Horizon)

Because this LEAP runs 2.4 years, near-term gamma/implied-move data (built for weeks-to-months) can only frame the first leg of the journey. Combined with the fundamental catalysts above:

📈 Bull Case (30% probability)

Target: Sustained move above $340, potentially toward $400-470 by 2028

  • Meta's Vera Rubin deployment ramps on schedule starting early 2027, Microsoft's NJ capacity scales, and the >4 GW power target is hit on time
  • Core AI-cloud ARR compounds well beyond the current $1.92B, and the $7-9B 2026 run-rate guide proves conservative
  • Financing continues on favorable terms with no dilution shock
  • Street targets get revised up past today's $380 high mark as the $46B Microsoft+Meta backlog converts to recognized revenue
  • If this happens, a long-call holder of today's cross profits handsomely (breakeven ≈$464); a short-call writer would be assigned or forced to roll/buy back at a loss

🎯 Base Case (45% probability)

Target: Choppy $150-300 range over the next 12-18 months, resolving toward guidance-dependent levels by 2028

  • Q2 earnings (July 29) come in roughly in-line with the $7-9B run-rate guide, no major surprise either way
  • Capex execution is real but bumpy — some site/power delays are normal for a company scaling this fast
  • Stock continues swinging within the gamma-mapped $180-$240 band near-term, occasionally testing $270-300 on good news and $150-180 on scares
  • This is the scenario the strong put-gamma tilt (26.5B put vs 13.3B call, total chain) currently reflects — a market still hedging two-way risk

📉 Bear Case (25% probability)

Target: Below $150, testing the $89.95 YTD starting price on a bad outcome

  • Capex ($20-25B planned) outruns financing capacity, forcing a dilutive raise or a spending pullback
  • Meta or Microsoft delays, renegotiates, or scales back capacity commitments (customer concentration risk realized)
  • Broader AI-capex-cycle wobble hits every "neocloud" name simultaneously
  • If this happens, a long-call holder of today's cross loses the full $18.6M premium at expiration; a short-call writer profits by keeping premium collected

The honest bottom line on probabilities: we can describe scenarios, but we cannot tell you which side of today's $18.6M cross is positioned for which outcome. That is the central, un-resolvable fact of this trade.

🎭 How Different Traders Should Read This

🎰 YOLO Trader

This LEAP print is not your setup. There's no short-term catalyst to trade around (earnings isn't until July 29, and even that barely dents a Dec-2028 option's price), and you can't tell which side "won" the cross. If you want NBIS convexity, the near-term gamma map ($180 support / $240 resistance) and this week's implied move (±14.75% into July 10) are far more actionable than trying to piggyback a 2.4-year block trade you can't even direction-tag.

📈 Swing Trader

Ignore the LEAP entirely for swing purposes — use the gamma levels instead. $210/$200 are your near-term supports on weakness; $240 is the first real overhead resistance on a bounce. The July 29 earnings date is 23 days out and will move the stock more than this cross ever will over your holding period.

💰 Premium Collector

If the "premium seller" read is correct (a long-term holder writing a deep 54%-OTM call for income), that's a legitimate, low-probability-of-assignment overwrite strategy you could mirror on a smaller scale with your own NBIS shares — sell a far-dated, far-OTM call to collect premium while retaining most of the upside. Just know the strike ($340) and tenor (2.4 years) here are extreme; a shorter-dated, closer-to-the-money version is the more standard retail version of this trade.

🌱 Beginner

The single most important lesson from this trade: not every big options print is a "bet." A block cross means two parties already agreed on a price privately — it's a negotiated transaction, not a signal that "smart money thinks NBIS goes up." Before you ever try to copy a headline options trade, learn to check whether it printed as a cross (no aggressor, unprovable direction) versus a lit sweep (someone actively buying/selling on the open market, which the tape can tell you about).

⚠️ Risk Factors & Honest Limits

What the tape genuinely cannot tell us:

  • ❌ Whether the cross was buyer-initiated or seller-initiated — no aggressor exists in a negotiated block
  • ❌ The counterparty's identity, broker/MMID, or whether either side already holds NBIS stock, other options, or a hedge
  • ❌ Whether this is one leg of a larger structure (a collar, a financing package, a delta-hedged position) invisible on the options tape alone
  • ❌ Whether tomorrow's OI increase (if it happens) represents a genuinely new directional bet or simply new contracts serving an already-hedged, market-neutral package

What could go wrong with the underlying business, regardless of this trade:

  • 💸 Capital intensity: $20-25B of planned 2026 capex vastly exceeds current revenue and ≈$9.3B cash on hand; execution depends on continued access to affordable capital, and the 2031/2033 convertible notes add leverage and potential dilution.
  • 🤝 Customer concentration: A large share of the contracted backlog sits with just two counterparties — Meta and Microsoft. Renegotiation, delay, or in-sourcing by either would materially dent the multi-year revenue ramp this LEAP is implicitly betting on.
  • 🏗️ Execution/power/supply-chain risk: Hitting >4 GW of contracted power by end-2026 and delivering Vera Rubin capacity in early 2027 depends on power availability, construction timelines, and GPU supply — any of which can slip.
  • 📊 Valuation dispersion: A $120-$380 analyst target range on the same stock says the Street itself isn't sure how much of the backlog converts to durable earnings. A +143% YTD run (with a recent -20% pullback already underway) leaves little margin for a stumble.
  • 🌀 Extreme volatility: 105.4% annualized volatility and a -32% max drawdown already this year mean NBIS can move double digits on no particular news. A 2.4-year hold will pass through many such swings.

🎯 The Bottom Line

Here's the deal: $18.6 million just changed hands on a Nebius call option that doesn't expire until December 2028 — and because it printed as a negotiated block cross at the exact midpoint of the market, we genuinely cannot tell you whether the buyer or the seller of that trade is the one who thinks NBIS goes higher. That's not us being cagey; it's the honest limit of what the options tape can prove.

What we can say with confidence:

  • ✅ Size (1,500 contracts) massively exceeds the strike's prior open interest (60) — this is very likely a fresh open, and tomorrow's ≈06:30 ET OI snapshot should show that (expect OI near 1,560)
  • ✅ The strike ($340) sits ≈54% above today's spot, and the tenor (2.4 years) brackets exactly the window in which Nebius's Nvidia, Meta, and Microsoft mega-deals are supposed to convert from signed contracts into recognized revenue
  • ✅ A long-call breakeven here (≈$464) sits above even the most bullish current Street target ($380) — whoever benefits from NBIS rallying needs the stock to outrun today's analyst consensus, not just meet it
  • ✅ This is a multi-year positioning trade, not a signal about what happens this week or even around the July 29 earnings print

If you're watching NBIS:

  • 👀 Mark July 29, 2026 for Q2 earnings — that's the next real catalyst, and it will move the stock far more than today's cross
  • 📊 Near-term, use the gamma-mapped $210/$200/$180 supports and $220/$240 resistance for actual short-term levels — not the LEAP strike
  • ✅ Next-day OPRA OI confirmed the $340 call strike OPENED (60 → 1,510, +1,450 ≈ the 1,500 block) — but as predicted, it does not resolve who is right on direction (a cross has no aggressor)

This is a marathon story, not a headline to chase. The AI-infrastructure buildout Nebius is riding is real and well-documented — but a single cross trade, however large, doesn't tell you how it ends. Do your own homework on the fundamentals, and don't mistake "someone traded a large size" for "someone knows something you don't."

Disclaimer: Options trading involves substantial risk of loss and is not suitable for all investors. This analysis is for educational purposes only and is not financial advice. Past performance does not guarantee future results. This specific trade printed as a block cross with no identifiable aggressor — its direction is genuinely unprovable from public market data, and nothing in this article should be read as a claim that the trade is bullish or bearish. Always do your own research and consider consulting a licensed financial advisor before trading, especially with long-dated, illiquid LEAP options.

About Nebius Group N.V.: Nebius Group is a vertically integrated AI-cloud and high-performance-computing infrastructure provider headquartered in Amsterdam, with a market cap of ≈$56 billion, classified in the software/AI-infrastructure ("neocloud") space.

Last updated: 2026-07-07 — open/close RESOLVED via next-day OPRA OI. Dec-15-2028 $340 Call 60 → 1,510 (+1,450 ≈ the 1,500 block) = OPENED. Block cross: direction remains unprovable. (July 3 was a full market holiday; baseline snapshot July 2 end-of-day, resolving snapshot July 6.)

The Options Desk tracks the move options price into every US earnings report the week of Sep 14, next to how much each stock has actually moved on its past prints — plus the SPY, QQQ and IWM expected ranges and the gamma walls that box them in.