🚀 2026 Q2 13F Holdings Deep Dive: Where Is Smart Money Betting?

[Miser][Smile]Fellow Tigers, Q2 13F filings are fully disclosed, and this quarter is packed with signals—Buffett's third massive $Alphabet(GOOG)$ purchase, Tepper loading up on Magnificent 7 while dumping memory stocks, and $SpaceX(SPCX)$ getting swarmed by hedge funds post-IPO.

📌 One-Sentence Summary in Adavance

Q2 smart-money rebalancing theme: loading AI giants (especially Alphabet), embracing the SpaceX space narrative, taking profits in memory/semis, concentrating China ADRs into AI-core names, and quant giants vs. discretionary managers taking opposite macro views.

Buffett's third Google purchase, Tepper dumping memory while buying Magnificent 7, hedge funds swarming SpaceX, Renaissance buying Meta and selling Micron, and Soros loading SMCI and utilities—together, these five moves map out where institutional capital is flowing for H2 2026.

[Allin][USD]💬 Community Discussion: Do your Q2 holdings overlap with these legends? Did you participate in SpaceX post-IPO? Are you bullish on Intel's turnaround? Share your thoughts below!

I. Buffett's "Third" Major $Alphabet(GOOG)$ Purchase—Alphabet Becomes Berkshire's #3 Holding

Berkshire added roughly 48.1 million shares of Alphabet in Q2, an 83% increase, bringing total holdings to ~106 million shares worth $36.6–$37.8 billion. Alphabet is now Berkshire's third-largest position, behind only Apple (~$69.7B) and American Express (~$51.9B).

Notably, about 60% (~$10B) of this addition came via a private placement, with the rest from open-market purchases. Alphabet raised $85 billion earlier this year for AI expansion, and Berkshire participated heavily in that placement.

Buffett's Alphabet accumulation trajectory:

  • Q3 2025: New position of ~17.8M shares (~$209 avg)

  • Q1 2026: Added 204%, ~36.4M shares (~$320 avg)

  • Q2 2026: Added another 45.2%, ~24.5M shares (~$359 avg)

Buying from $209 to $359 is rare for Buffett. It signals deep conviction in Alphabet's AI ecosystem flywheel (Gemini + Google Cloud + YouTube + search moat).

Rebalancing elsewhere:

Cash reserves dropped ~8% to $365.5 billion, indicating Buffett (and successor Greg Abel) are converting cash into equities.

II. David Tepper Q2 Rebalancing: Loading Magnificent 7, Dumping Memory

David Tepper's Appaloosa Management reported 27 positions worth $7.73 billion (up from $5.93B in Q1), yet trimmed holdings from 31 to 27 names—concentration is increasing.

🔵 Additions: Tech Giants + China Core Assets

Ticker

Action

Key Data

$Amazon.com(AMZN)$

+15.74% (+680K shares)

5M shares, ~$1.19B, remains #1 holding

$Alphabet(GOOG)$

Added

~$661M, ~8.5% of portfolio

$Meta Platforms, Inc.(META)$

+55%

~$380M

$Taiwan Semiconductor Manufacturing(TSM)$

+24%

~$788M

$Uber(UBER)$

+21%

~$555M

$Baidu(BIDU)$

+87%

~$148M

$Vistra Energy Corp.(VST)$

+10%

🔴 Reductions / Exits: Memory + China E-Commerce + Defense

  • Micron (MU): Cut 41% (1.65M → 975K shares). Dollar value still ~$1.13B (#2 holding) due to price appreciation, but Tepper is clearly taking profits in memory.

  • Fully exited 12 positions: SanDisk, Corning, PDD Holdings, JD.com, L3Harris, RTX, Ball Corp, Microsoft, KWEB (China Internet ETF), UnitedHealth, Lyft, Deutsche Bank

  • Major trims: Whirlpool (-63%), Qualcomm (-50%), Alibaba (-42%), AMD (-11%)

Tepper's logic is clear: memory cycle peaking → take profits; China e-commerce uncertainty rising → exit PDD/JD, keep only Baidu; defense stocks fully liquidated.

🆕 New Positions: SpaceX, $Boeing(BA)$ , $CoreWeave, Inc.(CRWV)$

New buys include SpaceX (225K shares, ~$38.4M), Boeing (80M shares, ~$173M), American Airlines (~$136M), CoreWeave (~$107M, AI infrastructure), and Broadcom (~$56.7M).

Note: Tepper also bought Apple puts (notional ~$242M) and Berkshire puts (~$12.5M)—likely hedges or expressions of near-term caution.

III. $SpaceX(SPCX)$ Post-IPO: Hedge Funds Pile In

SpaceX listed on Nasdaq on June 12, 2026 (IPO at $135). Q2 13F was its first appearance, and nearly every major hedge fund showed up.

Biggest Holders:

Firm

Shares

Value

% of US Equity Book

D1 Capital (Dan Sundheim)

>126M

>$21.5B

61%

Darsana Capital (Anand Desai)

>101M

>$17.3B

78%

Coatue (Philippe Laffont)

~18.6M

#4 holding

Millennium

~17.6M

#3 holding

Balyasny

~3.4M

#4 holding

Also disclosing SpaceX: ExodusPoint, Schonfeld, Element Capital, Tudor, Tiger Global, and more.

Important caveat: Many of these "new" SpaceX positions are legacy private investments converted to reportable securities post-IPO, not all fresh Q2 open-market buys. But D1 and Darsana's concentration—61% and 78% of their US books—is still staggering.

Baillie Gifford also shows SpaceX as its largest holding (8.66%), with total portfolio value rising from $98B to $110B.

IV. Other Top-Tier Firms: Q2 Full Picture

🎯 Steven Cohen (Point72): Semis + Tech Focus, Book Expands to $90.7B

Point72's 13F book grew 16.2% to $90.68B, with top-10 concentration at ~11.21%.

Cohen's direction: heavy semis (CRDO, ASML, AMD, TSM) + new Google position, hedged with SPY puts.

🎯 Ken Griffin (Citadel): $875B Behemoth, 22.67% Top-10 Concentration

Citadel Advisors reported $875.1 billion in 13F securities across 5,963 positions. Top-10 concentration: 22.67%.

Critical note: Citadel's 13F combines hedge fund + market-making inventory. The latter is temporary inventory, not long-term investment. Long-bias remains concentrated in SPY, QQQ, MSFT, AMZN, AAPL—mega-cap liquidity plays.

🎯 Renaissance Technologies: $72.6B, Buying META, Selling MU

Jim Simons' quant giant grew its book from $63.9B to $72.6B (+14%), holding 3,140–3,761 positions, top-10 at 12%.

  • Largest new position: $Amazon.com(AMZN)$ —2.29M shares, ~$546M

  • Biggest increase: Meta—estimated ~$1.08B added

  • Biggest reduction: Micron—estimated ~$1.46B sold

  • Exited: AppLovin (~$174M)

  • Q2 activity: 548 new positions, 1,084 increased, 1,502 reduced, 621 exited

Tech sector weight rose from 22% to 24%. As a pure quant fund, these moves reflect model signals rather than subjective views—but the memory sell-off aligns with Tepper's direction.

🎯 Two Sigma (John Overdeck): $1,381B, SPY Calls Quadrupled

Quant giant Two Sigma reported $1,381B across 3,963 positions, top-10 at just 8.28%.

  • Most aggressive move: SPY call position quadrupled to 546,800 shares—model is extremely bullish on the broad market

  • Top holdings: $Western Digital(WDC)$ (~1.1%), Nvidia, Apple

  • Estimated annualized return: +95.7%; YTD +24.8%

Two Sigma's model saw enough signals by end-June to justify a concentrated broad-market long. This directly contrasts with Point72's SPY putsquant giants and discretionary managers have opposite macro views.

🎯 Chase Coleman (Tiger Global): $24.0B, Slimmed to 46 Names

Tiger Global's book rose from $22.85B to $23.98B, but holdings slimmed from 54 to 46 positions.

  • Top 4: TSM, AMZN, NVDA, GOOGL

  • Q1 actions (context): Added heavily to NVDA, TSM, META, AVGO, Spotify, Coupa, AMAT, Zillow; exited Flutter, Veeva, Grab, Workday; trimmed AMZN, MSFT

Tiger continues its concentrated tech-giant strategy—top 4 cover AI chips, cloud, and search.

🎯 Philippe Laffont (Coatue): $48.6B, SpaceX Enters Top 4

Coatue's book surged from $29.0B to $48.6B (+67%), with 66 positions and 61.86% top-10 concentration—extremely concentrated.

  • Top 4: TSM, $Lam Research(LRCX)$ , Micron (MU), SpaceX

  • Exited 16 positions: Adobe, AMD, ARM Holdings, Axon Enterprise, Snowflake, Oracle, Marvell, Moderna, and more

  • Q1 top 5: TSM, GE Vernova, LRCX, AMAT, Broadcom

Laffont is executing an aggressive pivot: exiting software (Adobe, Snowflake, Oracle) while making SpaceX a top-4 holding. Coatue is rotating from "pure AI software" toward "AI infrastructure + space."

🎯 Dan Loeb (Third Point): $4.65B, WBD Becomes #1

Activist Dan Loeb's book exploded from $2.08B to $4.65–4.70B, with 43 positions and 38.1% top-5 concentration.

  • #1 holding: $Warner Bros. Discovery(WBD)$ at 11.4% (~$533M)

  • #2–3: Amazon (8.9%), Alphabet (7.8%)

  • 16 new positions: Block ($194M), Keysight ($205M), Norfolk Southern ($189M), etc.

  • Major adds: TSM (+67%), Live Nation (+144%), Capital One (+489%)

  • Exited 11: Including Microsoft and partial PG&E from Q1

Loeb's playbook is classic event-driven: heavy WBD = distressed turnaround + activist pressure, plus Alphabet/Amazon for tech, TSM for AI chip manufacturing.

🎯 George Soros (Soros Fund Management): $8.14B, 87 New Positions

Soros Family Office reported $8.14B across 266 positions, top-10 at 25.16%.

Soros' playbook is intriguing: new SMCI (AI servers), Nebius (AI infrastructure), plus massive utility bets (AEP, Entergy) and data center REITs (Digital Realty). The logic: AI compute needs power, power needs infrastructure—a full supply chain bet.

🎯 Israel Englander (Millennium): $1,422B, 3,592 Positions

Millennium reported $1,422B across 3,592 positions—the largest multi-strategy book.

  • Top 4: IWM (Russell 2000) puts, IVV (S&P 500 ETF), NVDA puts, QQQ puts

Millennium's book is dominated by index/options hedging + long/short single names. All top 4 are puts—active risk management via derivatives. Notably, Millennium also disclosed ~17.6M SpaceX shares as its #3 holding.

🎯 Paul Singer (Elliott Management): $22.7B, Extremely Concentrated

Elliott reported $22.67B across only 29 positions, with 79.91% top-10 concentration.

  • Top 4: TFPM.TO (Triple Flag Precious Metals), Phillips 66 (PSX), QQQ puts, SU.TO (Suncor Energy)

Elliott's style unchanged: ultra-concentrated, event-driven, energy/materials bias, hedged with QQQ puts.

🎯 Bill Ackman (Pershing Square): Exited Google, Embracing Payments & Streaming

  • Book expanded from 10 to 14 positions, value rose to $19.47B

  • New: Visa, Mastercard, S&P Global, Netflix

  • Fully exited: Alphabet and UMG

  • Major adds: Uber, Microsoft, Howard Hughes, Restaurant Brands, Meta

Ackman and Buffett took opposite actions on Alphabet—one exited, one added 83%. Proof that 13F is reference only; the same stock can have completely different conclusions under different frameworks.

🎯 Greenwoods Asset Management: $2.19B, China + Google Dual Thesis

China's top USD fund Greenwoods disclosed 21–27 positions worth ~$2.19B, covering communication services, tech, and consumer discretionary.

  • Top 5 holdings:

    • Alphabet (GOOGL): 29.6% (#1, up from 21.7% in Q1)

    • Intel (INTC): 17.6% (up from 7.9%)

    • PDD Holdings: 16.7%

    • Full Truck Alliance (YMM): 14.7%

    • Futu (FUTU): 4.7%

Greenwoods' moves are highly representative: massively adding Google (21.7% → 29.6%) and Intel (7.9% → 17.6%) while maintaining core China positions (PDD, YMM, FUTU). This aligns perfectly with Buffett's Google accumulation and Tepper's Baidu addition—China's top firms are also betting on AI via US equities.

V. Key Takeaways for Community Investors

1. "AI + Space" Dual Theme Is Locked In

From Berkshire and Tepper to Baillie Gifford, D1, and Coatue—top firms agree: AI remains core, and commercial space (led by SpaceX) is emerging as the next major theme after AI. If the past two years were about "AI chips," the future may shift toward AI application monetization, energy supply, and the commercial space supply chain.

2. Memory/Semi Cycle Topping?

Tepper cut MU -41%, exited SanDisk; Renaissance sold ~$1.46B of Micron; Coatue trimmed MU. After massive semi runs, smart money is taking profits in hardware. Chasing semis at these levels requires more caution.

3. China ADRs Diverging

Tepper exited PDD and JD but added Baidu +87%; Greenwoods kept PDD/YMM/FUTU but doubled down on Google and Intel. Firms are not treating China as a monolith—they're picking AI-driven names with clear moats.

4. Buffett's "Third Purchase" Signal

Buffett's three consecutive quarters of Alphabet accumulation, including a massive private placement, is a very strong long-term signal. Buffett rarely chases highs; his conviction in Alphabet's AI moat and cash-flow generation is worth noting for value investors.

5. Severe Macro Disagreement Among Top Firms

  • Two Sigma: SPY calls quadrupled → extremely bullish broad market

  • Point72 / Third Point / Elliott: Heavy SPY/QQQ puts → hedging downside

  • Tepper: Apple puts + Berkshire puts → cautious on select mega-caps

Even top-tier firms have wildly different views on H2 direction. Retail investors should not blindly copy 13F moves without understanding each firm's framework.

6. 13F Limitations Remain

  • 45-day lag: Q2 holdings are as of June 30, disclosed August 14—firms may have already rebalanced

  • Long-only view: Shorts, true option costs, cash, bonds, and non-US equities are not disclosed

  • "New positions" may be private-to-public conversions: Many SpaceX "new buys" are legacy private stakes becoming reportable post-IPO


Disclaimer: This article is compiled from public 13F filings for informational purposes only and does not constitute investment advice. 13F data has a 45-day disclosure lag and excludes shorts, cash, and non-US holdings. Please use rationally.

# 💰Stocks to watch today?(17 August)

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Report

Comment2

  • Top
  • Latest
  • 苏36
    ·08-18 21:43
    TOP
    The Q2 13F season reveals a clear message: institutional money is not abandoning AI—it is becoming more selective.

    Berkshire’s 83% increase in Alphabet, taking the position to roughly $38 billion, is perhaps the strongest vote of confidence in Google’s AI ecosystem.  Meanwhile, Tepper is rotating away from memory names such as Micron while adding Amazon, Meta, Alphabet and TSMC, suggesting investors may be taking profits after the semiconductor rally.

    The SpaceX story is equally important. Its IPO has brought massive institutional exposure into the public market, although some reported “new” positions are simply legacy private holdings becoming reportable.

    My takeaway: the next phase of the AI trade may shift from chips toward platforms, infrastructure, power and space. But 13Fs are snapshots, not real-time signals—investors should follow the thesis, not blindly copy the trades.

    @AI_FocusedTrader [你懂的]

    Reply
    Report
  • 我觉得看这份13F最重要的不是“抄谁的作业”,而是区分真正的新资金选择存量仓位被动显性化

    比如SpaceX很典型。IPO以后大量原本不可见的私募持仓变成13F里的公开证券,看起来像“对冲基金集体蜂拥买入”,但其中一部分其实只是旧仓第一次被市场看见,并不等于这些基金在二季度突然一致看多。所以单纯拿13F新增持仓判断边际资金流向,很容易高估共识强度。

    反而更有意思的是Tepper的动作:一边加Amazon、Google、Meta、TSM,一边减Micron、退出SanDisk。说明他并不是简单押注“AI全部上涨”,而是在AI产业链内部做取舍——更愿意拿平台、算力和确定性现金流,降低存储这种强周期、高beta环节的敞口。

    这可能也是这份13F真正值得看的地方:聪明钱并没有离开AI,只是在从“只要沾AI就买”,逐渐进入比较商业模式、盈利兑现和估值性价比的阶段。

    另外13F本身有45天滞后,而且看不到完整的空头、衍生品和盘中调整,所以最适合拿来研究机构的长期资产偏好,而不是直接当成买入信号。

    Reply
    Report