Crypto KOL Community & Market Trends

ETH vs. Russell 2000
Two signals have appeared ahead of major crypto rallies before:
• Russell 2000 breaks out
• ISM Manufacturing pushes above 55
In 2016, the setup appeared before ETH surged from $10 to $1,400.
In 2020, it happened again before ETH climbed from $88 to $4,800.
Now in 2026, the Russell has broken its multi-year range while ISM has reached 55.6, its highest level in four years.
Both signals are flashing again.
ETH has already bounced from its $1,505 low.
History doesn't guarantee the future, but this setup deserves attention.
#ETH #Crypto

A whale spent 2 million USDC to buy 518 million $PUMP in October last year at $0.00386. 1 hour ago, he sold all of his PUMP holdings, which he had held for 10 months, at $0.00274. He received SOL worth $1.42 million, incurring a loss of $580,000 (-29%). Address:

30 billion remains committed to Intel, 20 billion heavily invested in SpaceX! NVIDIA’s 8 stealth holdings:
Increased Holdings
1. SpaceX $SPCX
newly opened a position of approximately 123 million shares, with an ending market value of approximately $20.98 billion, suddenly becoming the second-largest holding (accounting for approximately 33.1%). SpaceX possesses the world’s leading reusable rocket technology and the Starlink satellite network, with strong competitive moats in space-based internet, commercial launches, and future space infrastructure. It also forms strategic synergies with NVIDIA’s AI computing ecosystem.
2. Intel $INTC
continued to rank as the largest holding (accounting for approximately 47.3%). Intel is a major global semiconductor manufacturer. NVIDIA has deep manufacturing and data center cooperation with Intel, and the two companies have formed a long-term strategic alignment in advanced process technology and AI chip foundry services.
3. CoreWeave $CRWV
holdings remained unchanged, with a market value of approximately $4.7 billion. CoreWeave is an emerging company focused on AI cloud computing, providing large-scale GPU cloud computing services. It directly benefits from the boom in generative AI and is highly tied to NVIDIA’s chip ecosystem.
4. Coherent $COHR
holdings remained unchanged, with a market value of approximately $3.07 billion. Coherent is a leader in optical communications and laser technology. Its products are widely used in data center optical modules and AI infrastructure, benefiting from the surge in demand for high-speed optical interconnects.
5. Nokia $NOK
holdings remained unchanged, with a market value of approximately $2.21 billion. Nokia has deep expertise in 5G/6G communications equipment and network infrastructure, and its technology is highly relevant to AI edge computing and smart network upgrades.
6. Synopsys $SNPS
holdings remained unchanged, with a market value of approximately $2.15 billion. Synopsys is a global leader in electronic design automation (EDA) software. Nearly all advanced chip designs rely on its tools, and the increasing complexity of AI chip design directly supports its long-term demand.
7. Nebius $NBIS
holdings remained unchanged, with a market value of approximately $330 million. Nebius focuses on AI cloud services and infrastructure. As an emerging computing power provider, it complements NVIDIA’s ecosystem strategy for deploying AI in the cloud.
8. Generate Biomedicines $GENB
holdings remained unchanged, with a market value of approximately $14.06 million. Generate Biomedicines uses generative AI to accelerate protein and drug design. It is a representative company in the AI and biopharmaceutical intersection, reflecting NVIDIA’s long-term optimism about AI applications in life sciences.
NVIDIA is so lazy—it only bought SpaceX and didn’t change anything else.

Crazy! It splurged $26 billion in a single quarter—did Saudi Arabia’s sovereign wealth fund really bet 70% of its holdings on Musk?

Reduced holdings
1. Allurion Technologies options warrants $ALUR
Its core product is the Allurion Balloon, currently the only swallowable, non-surgical intragastric balloon in the world. Combined with AI-powered remote monitoring, it helps overweight and obese patients lose 10–15% of their body weight in approximately 16 weeks. The related warrants were fully liquidated, further focusing the portfolio on core long-term strategic assets and reducing peripheral, non-core exposure.

Increased holdings
2. SpaceX $SPCX
Initiated a new position of approximately 154.1 million shares, worth approximately $26.34 billion and accounting for nearly 70% of the entire disclosed U.S. stock portfolio. As a global leader in commercial spaceflight, SpaceX controls reusable rocket technology and the Starlink satellite internet network. It possesses unique monopolistic advantages in low-Earth-orbit communications, space transportation, and the future space economy, offering enormous growth potential and a high degree of alignment with Saudi Arabia’s diversification strategy.
3. Uber $UBER
The number of shares remained unchanged at approximately 72.84 million, worth approximately $5.26 billion. Uber has powerful network effects and scale advantages in global mobility and delivery. Its platform business continues to expand, profitability has improved significantly, and its long-term growth outlook is highly predictable.
4. Electronic Arts $EA
The number of shares remained unchanged at approximately 24.81 million, worth approximately $5.09 billion. Electronic Arts is one of the world’s top game publishers, with classic IP such as FIFA and Battlefield. Its content ecosystem is solid, its subscription and livestreaming businesses have strong growth potential, and its cash flow is strengthening.
5. Lucid $LCID
The number of shares remained unchanged at approximately 177.1 million, worth approximately $1.18 billion. As a premium electric vehicle brand, Lucid has leading technology and deep cooperation with Saudi Arabia. It benefits from the global shift toward electrification and localized production initiatives, giving it significant long-term strategic value.
6. Claritev $CTEV
The number of shares remained unchanged at approximately 1.28 million, worth approximately $43.7 million. As a healthcare data and management services company, it has stable operations and niche advantages in healthcare digitalization. It was retained as a defensive allocation within the portfolio.
A massive bet on the future!

U.S. Treasury yields hit a 19-year high, yet BTC rose—confused again, huh?

The 30-year U.S. Treasury yield: 5.29%.

The highest since 2007.

The textbook tells us: Long-term yields surge → borrowing costs rise → valuations are suppressed → risk assets plunge.

By that logic, Bitcoin should be bleeding heavily today.

But it rebounded.

On August 17, BTC rebounded from a low of $62,714 to $64,360.

Has the market broken down?

Or is your analytical framework outdated?

First, let’s look at what happened today.

U.S. Treasuries faced a new wave of selling, with the 30-year Treasury yield rising 3 basis points on Monday to 5.29%, approaching the 5.44% high reached at the beginning of the 2008 financial crisis.

Meanwhile:

The probability that the Federal Reserve will keep rates unchanged in September rose to 69%

The 2-year Treasury yield has fallen by approximately 20 basis points since July 23

Bitcoin rebounded to $64,360

Short-term rates are falling while long-term rates are surging.

The yield curve is undergoing “bear steepening.”

The market is not worried about rate hikes in the short term, but is absolutely terrified of them in the long term.

Afraid of what?

Afraid of the Federal Reserve. Afraid of the Treasury. Afraid that neither institution dares to do the right thing.

Nohshad Shah, head of fixed-income sales at Castle Securities, delivered a harsh assessment:

“This reflects the market’s belief that, when faced with difficult choices, both the Federal Reserve and fiscal authorities tend to choose the easier path.”

What does that mean?

With inflation above the Federal Reserve’s target for five consecutive years, the Fed does not dare to hike rates again—the probability of a rate hike plunged from 75% in late July to 33%.

With debt exceeding $40 trillion, the Treasury does not dare to cut spending—it can only continue issuing debt and borrowing money.

Policymakers will always choose “the easier path.”

And the market is paying the price for this “ease.”

The numbers do not lie.

Hartnett, BofA’s chief investment strategist, identified the core issue:

Over the past 12 months, U.S. interest expenses on debt have reached $1.4 trillion, approaching the point of surpassing Social Security to become the federal government’s largest single expenditure.

The 30-year Treasury was issued last week at a yield of 5.126%, a 25-year high.

Hartnett’s exact words:

“U.S. stocks hit an all-time high on the same day that Treasuries were issued at their highest yield in 25 years—this is reality.”

The reality is that it is becoming increasingly expensive for the government to borrow money and increasingly difficult to repay it.

So why is Bitcoin rising?

Because the market has never been pricing in “higher rates.”

The market is pricing in “the Federal Reserve not daring to hike rates again.”

When Castle Securities says the Fed is “unwilling to tighten monetary policy further,” the bond market hears: “Inflation will continue.”

When the Treasury continues issuing debt at rates above 5%, the bond market hears: “Debt will continue to expand.”

When these two messages combine, smart money hears:

“The dollar cannot be trusted over the long term.”

Jim Bianco put it even more bluntly:

“The time bond traders can stop panicking is when the Federal Reserve starts panicking.”

In other words, as long as the Fed continues pretending to stay calm and takes “the easier path,” long-term yields will not have peaked.

And Bitcoin?

It does not care whether the Fed is panicking. It only cares whether the dollar is depreciating.

So today’s counterintuitive market action is actually very clear:

The bond market is saying, “I’m afraid of inflation”—so the 30-year yield surged to 5.29%.

The crypto market is saying, “I’m even more afraid of the dollar”—so Bitcoin rebounded to $64,360.

These two statements do not contradict each other.

They are saying the same thing:

When policymakers always choose “the easier path,” the true safe-haven assets are not Treasuries, but assets that are beyond the control of central banks.

A 5.29% 30-year Treasury yield is not fear of Fed rate hikes.

It is fear of the Fed’s incompetence.

When the market realizes that neither the Federal Reserve nor the Treasury has the courage to make difficult decisions—

Capital will find its own way out.

And today’s Bitcoin is a microcosm of that way out.

What do you think—can this BTC rebound continue?

Disclaimer: The above is purely a personal opinion and does not constitute investment advice. DYOR. #Gate事件积分系统上线 #我的七夕交易分享 #SpaceX股价反弹40% $BTC $ETH $SOL

Gold prices rose 8%, but the “Hermès of gold” fell 46%: Capital markets are redefining what “valuable” means

Since August, COMEX gold has risen 8.2%.

What about Laopu Gold? It rose just 2.7% over the same period, and fell another 5.2% over the past week.

Gold prices are back, but Laopu isn’t.

Do you know anyone like this—watching gold prices rebound, assuming gold stocks would surely rise too, rushing in to buy the dip on Laopu, and ending up losing 5% in a week?

Capital markets are using real money to tell you one thing: Laopu has never been a gold stock.

In July 2025, Laopu Gold’s share price surged to a record high of HK$1,108. At its peak, its market capitalization was nearly HK$188.3 billion.

On August 14, 2026, its share price fell to HK$333.6.

From HK$1,108 to HK$333, that was a 68% drop, wiping out more than HK$120 billion in market value.

Year to date, the COMEX gold front-month contract has risen 1.3%, and is up 8.9% this quarter.

Laopu Gold is down 46.02% year to date.

Gold is rising, while Laopu is falling. Gold prices are back, but Laopu isn’t.

Why is this happening?

The answer lies in Laopu’s business model.

Laopu doesn’t sell gold. It sells traditional craftsmanship, Chinese aesthetics, a storefront in SKP, and the sense of status that comes from being neighbors with Hermès.

Its gold jewelry is not sold by weight, but by the piece at a fixed price. Traditional craftsmanship, filigree engraving, blue enamel, and hollowwork—combined with its reputation as the “Hermès of the gold industry”—give Laopu’s products a premium generally equivalent to around three times the price of the raw gold.

What you buy at Laopu is a luxury jewelry item, not a gold asset.

When you take this “luxury item” to be recycled—

You pay a threefold premium when buying it, but at recycling, it is valued only at “the market price minus ¥100 per gram.”

One consumer took her Laopu necklace to ask about its recycling price. The gold price was lower when she bought it, and gold had risen again over the past year. By common sense, selling it should at least have allowed her to avoid a loss. But the price quoted by Laopu was roughly the same as what she had paid for it.

The purchase follows luxury-goods logic; the exit follows gold logic.

The gap between the two is the root cause of the share-price collapse.

More brutally, Laopu’s luxury positioning is now turning against it.

From 2025 to January 2026, gold prices rose from $2,600 to nearly $5,600. Laopu chose instead to aggressively stockpile gold, without carrying out any hedging. Founder Xu Gaoming publicly stated: “Once you choose to hedge, it means abandoning the original aspiration of building a world-class brand.”

As of the end of 2025, the company’s inventory balance was approximately ¥16 billion, nearly triple the amount at the end of the previous year. After gold prices retreated from their January peak, this high-priced inventory became a looming impairment risk.

The former “king of queues” has now begun offering large-scale discounts. Spend-and-save promotions are appearing one after another at Beijing SKP, Shanghai IFC, and Guangzhou Taikoo Hui, while scalpers and proxy buyers can get prices down to 82% of the original price.

Consumers who were still saying a few months ago, “If you don’t buy now, it’ll be more expensive later,” are now saying, “I lost ¥5,000 within a week of buying it.”

Here’s the painful truth—

It’s like buying the “Hermès of the crypto sector.”

No matter how attractive the narrative or how strong the consensus, the market will ultimately ask: What exactly are your underlying assets? When liquidity exits, what does the market use to price you?

Laopu’s foundation is gold, but the market was betting on the dream of a “Chinese Hermès.”

Once the dream faded, its valuation fell from 50x PE to 12x, leaving it barely different from traditional gold retailers such as Chow Tai Fook and Lao Feng Xiang.

The market has finally figured it out—when a luxury-goods bubble bursts, even gold cannot save you.

Gold rises, Laopu falls.

It’s not that the market was wrong. The market has finally seen clearly:

Once a product is priced as a luxury good, it no longer enjoys the β of a commodity.

And every bit of brand premium you pay when buying it will, at the moment of exit, be treated by the market as—

A bubble.#Gate事件积分系统上线 #我的七夕交易分享 #股票交易分享挑战 $BTC $XAU $XAUT

Dusk is moving beyond just building tech.
The OpenDusk initiative and treasury vote show the community becoming more involved in how the ecosystem’s resources are managed.
For me, that governance layer is just as interesting as the tech itself.
$DUSK #dusk @DuskFoundation

Last time when SMA 20 was broken $ETH surged more than 62% against $BTC

$BTC is still stuck between $62K support and $67K resistance.
Momentum indicators are improving, but BTC needs a clean breakout first.
Above $67K looks bullish.

Lose $62K and $58K comes back into play.
For now, the range is everything.

$XRP really said f*** that lower low 😭
Now price is sitting right under the $1.164–$1.18 liquidity zone.
If that gets swept too…
I’m not saying $1.20+ is coming, but rn I’m definitely not fading this juicy setup.
Could give us one nasty relief rally before the sharp dump.
Degens might get served another plate imo 🍽️
#xrp #Ripple

At $148, SPCX—are you waiting for $200 before getting in?

First, the surface: After its IPO, it fell from 225 to 104, down over 50%, and everyone called it “the bursting of the space bubble.”

Over the past month, it violently rebounded from 104.83 to 148, reclaiming the IPO price of 135, up over 5.5% today, with 24-hour volume expanding. Volume-backed breakout of the downtrend line from the bottom, MA5/10 crossing upward, RSI neutral to strong, and the recovery wave underway.

First: SpaceX is genuinely scarce, not one of those altcoins you usually buy.

SpaceX was not publicly listed before, so retail investors could only watch helplessly. Now SPCX trades on Nasdaq, with perpetuals available too—you can finally buy one of humanity’s top-tier space assets.

Goldman Sachs just released a report: The global space economy is expected to reach $1.8 trillion by 2035. SpaceX holds three trump cards—launch services, Starlink, and orbital computing—and is the purest play in this sector.

On August 14—SpaceX completed the acquisition of Cursor, an AI programming tool, for approximately $60 billion, directly strengthening its AI strategy. Nvidia disclosed a position worth approximately $21 billion, while major funds such as Harvard are also increasing their holdings.

Second: The first earnings report on August 4 was explosive, but many people failed to understand it.

Revenue was $18k, up 92% year over year and above expectations. Starlink is already profitable, AI revenue is growing explosively, the order backlog stands at $47.5 billion, and cash reserves are approximately $100 billion.

Net loss narrowed to $541 million—but that was the result of spending on AI infrastructure, not losses in the core business.

Capital expenditures were $18.4 billion, with a large amount directed toward AI infrastructure.

Third: The August 20 unlock bomb—but the previous “rise instead of decline” already said it all.

On August 6, the first batch of major locked-up shares was released (911.5 million shares). The market expected a sharp drop, but what happened? The stock rose instead of falling.

The next key point is the release of approximately 319 million shares around August 20, representing a market value of approximately $45 billion at the current price. The market is worried about selling pressure, but the previous experience has already proven that there is sufficient capital on the demand side to absorb it.

The bull-bear showdown—you decide

On one side:

Goldman Sachs forecasts a $1.8 trillion space economy, with SpaceX as the core leader

The $60 billion acquisition of Cursor—the AI + space narrative has just begun

Nvidia’s $21 billion position + Harvard increasing its holdings—institutions are accumulating

The August 6 unlock resulted in a rise instead of a decline, proving strong absorption capacity

Rebounded from 104 to 148, with the trend having turned bullish

On the other side:

Another 319 million shares will be unlocked on August 20, with selling pressure uncertain

Capital expenditures of $18.4 billion—the spending rate is astonishing

Market value of approximately $1.85 trillion, with the valuation still extremely expensive

Beta above 6—it falls even harder whenever the broader market declines

Perpetual contracts carry funding-rate and liquidation risks

Key levels

Upside resistance: 150–155 → 160–170 → 225.64 (ATH)

Downside support: 140 (psychological level of the previous high) → 135 (IPO price) → 125–130

Trading strategy

Short-term traders:

Try a small long position around 148, with a stop-loss at 142–143 and a target of 155–160; reduce your position once it reaches that level. If it breaks below 140 on heavy volume, stay on the sidelines or take a small short position targeting 135.

Conservative players:

Wait for a pullback to 145–146 or stabilization around 140 before entering; do not chase higher prices. Alternatively, wait until the August 20 unlock bomb has been defused—if it holds above 135 after the unlock and quickly reclaims it, add aggressively and target 170+. If it crashes through 135 on heavy volume, wait for 120–130 before buying the dip.

Long-term believers:

Build positions in batches. SpaceX is one of humanity’s top-tier technology companies, and its long-term narrative of Starlink + Starship + space AI is unmatched.

This is not a meme coin; this is the “most hardcore asset” you can buy in your lifetime.

It fell from 225 to 104, down 50%, then rebounded to 148—when it was falling, you didn’t dare buy; once it rose, you chased higher prices. You are always losing money.

The August 20 unlock will be the true test.

When a top-tier asset is on sale, you don’t dare buy—are you really going to wait for it to return to 200 and then chase higher?

What is your SPCX cost basis?

At 148, do you dare get on board? #SpaceX超级股东曝光股价反弹40% #我的七夕交易分享 #股票交易分享挑战 $BTC $SPCX $TSLA

The film 《Bull Comes》 has surpassed 10 million yuan in box office revenue
The $Bull Comes token’s market cap has surpassed $40 million
ca：0xbeea1d618e533a387d941f58a7d4c9b7bd377777
Based on this trend, it looks like breaking $100 million won’t be a problem
Will it be the next $Coin You?
ca：0xe9337dde3dd9e97f1f45a56412767ce5098e7777

Earnings season is more or less over. It is clear who is stronger and who is weaker. Follow the companies with strong earnings, and it will be hard not to make money in the second half of the year!

Strong

1. TSMC $TSM
reported second-quarter revenue of NT$1,270.38 billion, up 36% year over year, and net income of NT$706.56 billion, up 77.4% year over year. Both revenue and profit posted strong growth. The 2nm process has begun ramping into mass production, while demand for 3nm and 5nm remains extremely strong. Third-quarter revenue guidance reached $44.6 billion—$45.8 billion, and the company raised its full-year 2026 revenue growth forecast in U.S. dollar terms to above 40%. Continued demand for AI chips is keeping advanced-process and advanced-packaging capacity tight, making this the most important confirmation of the current AI semiconductor upcycle.
2. ASML $ASML
reported second-quarter sales of €9.33 billion and net income of €2.92 billion. Order demand remained strong, and the company raised its full-year 2026 sales guidance to €43 billion—€45 billion. Demand for high-end lithography equipment, including EUV and High-NA EUV, continues to grow, while capacity expansion for AI chips, advanced logic, and high-end memory continues to flow upstream to equipment suppliers.
3. Lam Research $LRCX
reported record quarterly revenue, with profit and cash flow increasing simultaneously. Next-quarter revenue guidance reached approximately $8.1 billion, significantly above market expectations. The expansion of AI data centers is driving continued increases in capital spending on advanced logic, HBM, and advanced packaging, and the company is benefiting directly from rising wafer-fabrication complexity.
4. KLA $KLAC
reported fourth-quarter revenue of $3.66 billion, at the high end of its guidance. GAAP EPS and Non-GAAP EPS also reached the high end of the guidance range. As AI chip manufacturing becomes increasingly complex, advanced processes, HBM, and advanced packaging all require more inspection and metrology equipment, giving the company relatively high earnings visibility.
5. Applied Materials $AMAT
reported third-quarter revenue of $9.12 billion, up 25% year over year, while adjusted EPS reached $3.50. Both exceeded market expectations. Fourth-quarter revenue guidance was approximately $10.25 billion, also above market expectations. The company also raised its 2026 advanced-packaging revenue growth forecast to above 70%, indicating that AI chip capacity expansion has begun to significantly drive equipment demand.
6. AMD $AMD
reported second-quarter revenue of $11.54 billion, up 50% year over year. Data center revenue reached $6.72 billion, up more than 100% year over year and also above market expectations. Server CPU demand remained strong, while the MI series of AI accelerators began ramping rapidly. The company expects data center revenue to double again in 2027. Overall, the earnings report was clearly strong, with the core thesis being that the AI computing business is growing rapidly.
7. Microsoft $MSFT
reported fourth-quarter Azure revenue growth of 43% year over year, above market expectations, with the cloud business exceeding $100 billion in annualized revenue. The company’s commercial AI products continue to be monetized, with more than 30 million paid Copilot users. Cloud remaining performance obligations reached $678 billion, while the company continued to maintain strong free cash flow. Most importantly, Microsoft has begun demonstrating that massive AI capital investment can be converted into actual cloud revenue.
8. Amazon $AMZN
reported second-quarter revenue of $200.6 billion, up 20% year over year, while AWS revenue grew 37% year over year, its fastest growth rate in nearly 18 quarters. AWS annualized revenue reached approximately $169 billion. Its internally developed Trainium, Graviton, and other AI-related chip businesses continue to expand, with AI cloud demand and data center investment jointly driving growth.
9. Arista $ANET
reported second-quarter revenue exceeding $3 billion for the first time, up approximately 38% year over year. Demand for AI data center networking continued to grow rapidly, and the company raised its full-year revenue guidance. Orders for 800G products, AI cluster networks, and large cloud providers continue to increase. The more AI servers there are, the greater the demand for high-speed switches and networking equipment.
10. Palantir $PLTR
reported second-quarter revenue of $1.94 billion, up 93% year over year, while adjusted EPS reached $0.41. Its U.S. commercial and government businesses continued to grow rapidly. Third-quarter revenue guidance was above market expectations, and the company sharply raised its full-year 2026 revenue forecast from approximately $7.65 billion to $8.15 billion—$8.158 billion. AI software has moved from pilot programs into large-scale commercial deployment.
11. Nebius $NBIS
reported second-quarter revenue of $582 million, up approximately 454% year over year and above market expectations, while AI Cloud revenue nearly increased sixfold. The company signed four AI cloud contracts with average contract values exceeding $1 billion. Total customer contract value nearly quadrupled, and the company raised its power-capacity target for the end of 2026 to 5GW. Total customer commitments currently exceed $40 billion. The company even stated that, based on current demand, the computing capacity it plans to deploy in 2027 could sell out ahead of schedule. It is one of the fastest-growing companies in the AI computing infrastructure chain during this cycle.
12. CoreWeave $CRWV
reported second-quarter revenue of $2.58 billion, up more than 100% year over year. Remaining performance obligations reached $104.2 billion, while new customer commitments added during the quarter exceeded $25 billion. The company continued to raise its 2026 revenue and adjusted operating profit targets, while increasing its full-year capital expenditure forecast to $35 billion—$39 billion. Although investment is massive, the backlog exceeding $100 billion shows that demand for AI computing capacity still far exceeds supply.
13. Lumentum $LITE
reported fourth-quarter revenue of approximately $1.01 billion, up 109% year over year, while adjusted EPS increased 267% year over year. Next-quarter revenue guidance reached $1.225 billion—$1.275 billion, continuing to accelerate significantly. AI data centers are driving explosive demand for 800G, 1.6T high-speed optical interconnects, and related lasers, making optical communications one of the clearest high-growth segments in AI infrastructure.
14. Coherent $COHR
continued to report rapid fourth-quarter revenue growth, with demand for high-speed optical interconnects from AI data centers serving as the core growth driver. As AI cluster sizes continue to expand, copper interconnects are gradually being upgraded to high-speed optical connections, benefiting the company’s optical components, laser, and data center connectivity businesses.
15. Super Micro Computer $SMCI
reported fourth-quarter revenue of approximately $11.1 billion, up approximately 93% year over year, while adjusted EPS increased significantly year over year. Data center and AI server orders continued to rise. The company’s outlook for the next fiscal year is highly aggressive, with rapid growth in orders for AI servers, liquid-cooled servers, and GPU infrastructure, directly benefiting from the construction of large AI data centers.
16. SanDisk $SNDK
reported fourth-quarter revenue of approximately $8.97 billion, while data center revenue grew more than 400% year over year. Next-quarter revenue guidance reached $10.3 billion—$10.8 billion. The company has signed eight long-term supply agreements with six customers, with a total value of at least $93.9 billion. Part of its capacity has already been locked in for the next several years. Demand for SSDs from AI data centers continues to surge, and the memory cycle is clearly in a strong phase.
17. Intel $INTC
reported second-quarter revenue of $16.1 billion, up 25% year over year, marking unusually rapid growth not seen in years. Its data center business improved significantly. The 18A process, advanced packaging, and Foundry business continue to advance, while AI-related businesses posted noticeable growth. Although the company’s transformation remains uncertain, the earnings report itself shows clear improvement in both revenue and business mix.
18. Cisco $CSCO
reported fourth-quarter revenue of $17.25 billion, up 17.6% year over year and above market expectations. Full-year AI infrastructure orders reached $9.3 billion, including $4 billion in the fourth quarter alone. The company expects fiscal 2027 revenue of $72.2 billion—$73.4 billion, above market expectations, and expects AI infrastructure revenue to continue growing. The expansion of AI data center networks is now directly reflected in orders and revenue.
19. Vertiv $VRT
reported second-quarter revenue of $3.274 billion, up 24% year over year, while adjusted EPS grew 60% year over year and adjusted operating margin reached 22.6%. The company also raised its full-year revenue, profit, and cash flow guidance. As AI data center power consumption continues to rise, demand for UPS systems, power equipment, liquid cooling, and thermal management continues to increase, making Vertiv a core beneficiary of AI power infrastructure investment.
20. Seagate Technology $STX
reported fourth-quarter revenue of $3.63 billion, up approximately 49% year over year. Data center revenue reached $2.9 billion, up approximately 57% year over year. The company expects next-quarter revenue of approximately $4.1 billion. Demand from AI data centers for high-capacity hard drives and storage remains strong, as the explosion in data volumes is directly translating into storage demand.
21. Constellation Energy $CEG
continued to deliver stable second-quarter results and continued advancing long-term power purchase agreements with major technology companies. AI data centers are becoming an important source of incremental electricity demand in the United States. Stable, continuously operating nuclear assets are gaining greater long-term value, and the expansion of AI computing capacity has begun to flow through to the power sector.
22. Vistra $VST
maintained steady operations in the second quarter and improved future revenue visibility through long-term power agreements with major technology companies. AI data centers require large amounts of reliable electricity, and long-term power purchase agreements allow the company to convert AI-driven power demand into relatively stable cash flow.

Weak

23. SK hynix $SKHY
reported explosive year-over-year growth in second-quarter revenue and profit, while HBM demand remained extremely strong. However, quarterly performance did not fully meet the market’s previously very high expectations. The pace of HBM shipments, product pricing, and future supply expansion have become new areas of focus. This was not a poor performance; rather, the incremental improvement in the earnings report did not continue to exceed an already extremely high market expectation.
24. Google $GOOGL
reported 82% year-over-year growth in second-quarter Google Cloud revenue, with very strong demand for AI-related businesses. At the same time, capital expenditures on AI data centers, GPUs, and infrastructure continued to increase substantially. The company raised its full-year capital expenditure plan, putting noticeable pressure on cash flow. The core question has shifted from “Is there demand for AI?” to “How long will it take for massive AI investment to translate into higher profit and free cash flow?”
25. Meta $META
reported second-quarter revenue of $60.8 billion, up 28% year over year, with its advertising business remaining strong. However, net income fell 14% year over year, operating margin declined significantly, costs and expenses increased 55% year over year, and free cash flow plunged 91% to $784 million. The company also raised its 2026 capital expenditure forecast to $130 billion—$145 billion. AI investment is highly aggressive, but profit and cash flow realization are currently lagging far behind the pace of investment.
26. Qualcomm $QCOM
delivered overall growth in the second quarter, with its automotive and IoT businesses continuing to expand. However, handset chips remain its primary source of revenue, and its AI data center business has not yet achieved growth on the scale of AMD or NVIDIA. The company is actively transforming toward automotive, PC, IoT, and data center markets, but its new AI businesses have made only a limited contribution to near-term profit growth. As a result, the AI content of this earnings report is lower than that of the core AI computing companies.

Next week, $NVDA $MRVL $AVGO big ones are coming!

BTC is still ranging around 63,600 — the calm before the storm, or the frog slowly boiling in warm water?

First, the surface picture: it can't fall, but it can't rise either.

For the past 10 weeks, BTC has repeatedly churned within the $4,000 range between 62,000 and 66,000. On August 17, perpetuals were at 63,600, up a slight 1.1% over 24 hours, with the daily range shrinking to under $1,000 and trading volume falling to near lows seen in recent years. From the all-time high of 126,272, it has already been cut in half, falling a full 50%.

You open the chart and feel like you're looking at an electrocardiogram — a straight line. Volatility has compressed to an extreme, the Bollinger Bands have narrowed, and a major move is imminent.

First: everyone is selling, but the price hasn't collapsed.

Strategy continues to reduce its holdings, miners are selling, ETFs have seen consecutive net outflows (-50M to -130M on a single day in mid-August), the probability of the CLARITY Act passing has been lowered to 10%, the regulatory narrative has gone cold, and a hardware wallet data leak has added to security fears.

Does it all sound bearish? Then why hasn't BTC fallen below 60k?

Because those who needed to sell have already sold most of what they had. If bad news can't push the price down, that's the biggest bullish signal.

Second: macro is about to play its hand.

The Federal Reserve is maintaining interest rates at 3.5%-3.75%, with the July FOMC vote at 9-3 and 3 committee members leaning toward a rate hike. Core PCE has risen to 3%+, and inflation is stickier than you might think.

This Wednesday (August 19), the FOMC meeting minutes will be released, and the market will see the extent of internal disagreement over inflation and the policy path. If the minutes are hawkish: risk assets will come under pressure, and BTC could retest 62,000 or even 60,000. If they are dovish or neutral: support will be more effective, opening a window for a rebound.

Third: the technical picture has reached a crossroads.

A descending channel plus a symmetrical triangle are converging on the daily chart, with price fluctuating near the channel's lower boundary. 62,800-63,000 has been tested and bounced multiple times, but every time price reaches 64,500-65,000, it gets slapped back down.

RSI is neutral at 45-48, MACD is weak, Stochastic is moving downward, the moving averages are bearishly aligned, and the risk of a death cross is faintly visible. Trading volume continues to shrink — no one is willing to bet on a direction here.

Key levels:

Support: 62,800-63,000 (short term) → 62,400 → 60,000-61,000 (liquidity zone) → 59,000-53,000

Resistance: 63,900-64,500 → 65,000-65,500 → 67,000

You decide who wins the battle between bulls and bears.

On one side:

10 weeks of ranging + extreme volatility compression, with a major move imminent

STH MVRV at 0.94, with losses continuing for nearly 100 days and most underwater coins effectively flushed out

LTHs are still accumulating, supply concentration is rising, and smart money hasn't left

The bad news has been exhausted, and those who needed to sell have already sold

On the other side:

ETFs continue to see net outflows, and institutional demand is weak

Strategy is reducing its holdings, weakening the corporate Bitcoin accumulation narrative

The probability of the CLARITY Act passing is only 10%, and the regulatory narrative has gone cold

The Federal Reserve is hawkish, core PCE is rising, and liquidity remains finely balanced

August seasonality is historically weak

Trading strategy

Short-term traders:

Bearish approach: scale into small short positions at 63,900-64,500, stop-loss at 64,800-65,000, targets at 63,000-62,400

Bullish approach: go long with a small position after a pullback holds at 62,800-63,000, stop-loss below 62,400, targets at 63,900-64,500

Swing traders (waiting for breakout confirmation):

A high-volume daily close holding above 64,500 → chase longs, targets 67,000-70,000

A high-volume daily close below 62,800 → follow the move short, targets 60,000-61,000

Long-term believers:

Below 62,000 remains an accumulation zone, but don't deploy everything at once. A macro shift + ETF inflows are the real right-side signals.

Wednesday's FOMC minutes will determine the direction for the next two weeks.

The calm before the storm is often the most agonizing. Ten weeks of ranging has shaken out most retail traders, leaving only veteran retail traders and market makers.

On the day 64,500 breaks out, or the day 62,800 breaks down, you'll realize:

It wasn't that BTC had no direction; you were getting shaken out before every major move.

What's your BTC cost basis? Do you think this week's FOMC minutes will be hawkish or dovish? #我的七夕交易分享 #比特币变盘信号出现 #GateLaunchpool瓜分141万枚DOS $BTC $ETH $SOL

Gold at $4,400—would you dare chase it?

First, look at the surface: it took gold six months to fall from $5,500 to $4,000, but only one month to rebound from $4,000 to $4,400.

What happened over the past week? U.S. July retail sales unexpectedly fell 0.6%, consumer confidence declined for the first time in three months, and economic data weakened across the board. Market bets on a September rate hike plunged directly from 55% to 33%. The weaker dollar directly benefited gold.

First: Inflation cooled, but gold did not fall.

U.S. July CPI rose 3.4% year-on-year, versus 3.5% previously; core CPI rose 2.5%, versus 2.6% previously. PPI rose 4.7% year-on-year, versus 5.5% previously. After two consecutive months of cooling, the market directly interpreted this as meaning the Federal Reserve will most likely stand pat in September.

Inflation is down, the rate-hike probability is down, and the dollar is weaker—if gold does not rise, it would defy reason.

Second: The Strait of Hormuz is the real “gold catalyst.”

On August 10, Trump claimed that the United States had “100%” control of the Strait of Hormuz. Iran’s deputy foreign minister directly hit back: “The Strait of Hormuz belonged to Iran in the past, belongs to Iran now, and will belong to Iran in the future.” The 60-day U.S.-Iran ceasefire agreement expires today, and there are no signs of negotiations between the two sides.

Daily traffic through the strait plunged from 130 vessels before the conflict to 13, a 90% drop. Multiple oil tankers have been attacked, while some vessels have switched off their satellite transponders to pass through covertly.

If oil prices surge, inflation will return, making the Federal Reserve even less willing to cut rates. But gold? The greater the geopolitical turmoil, the stronger the safe-haven demand. Every additional day of confrontation in the Strait of Hormuz provides another layer of support for gold.

Third: Central banks are buying aggressively—this is the strongest floor.

The World Gold Council’s Q2 report: Global central banks made net gold purchases of 289 tons, up 62% year-on-year and marking the highest quarterly purchase volume in nearly four years. Global gold demand totaled $380 billion in the first half of the year, a record high.

China’s central bank has increased its holdings for 20 consecutive months, while Poland and other emerging markets are also stockpiling. 45% of surveyed central banks expect to continue increasing their gold reserves over the next year.

Fourth: Technical signals have emerged that you must take seriously.

Daily chart: After consolidating in July, gold broke upward, met resistance at $4,450, pulled back to $4,320-$4,350, and quickly recovered. It is currently fluctuating around $4,400. RSI is approximately 60-65, not overbought; MACD is positive but momentum is weakening. The key point is that gold broke above the 100-day moving average last week for the first time since April and is still holding near that level.

Weekly chart: After retreating from the January high, gold formed a base around $4,000. It has rebounded approximately 10% since August and is in the process of repairing its uptrend channel.

Bulls versus bears—you decide

On one side:

CPI and PPI have both cooled, leaving the September rate-hike probability at just 33%

The Dollar Index has fallen to around 99.5, while real interest rates are declining

The confrontation in the Strait of Hormuz continues, with strong geopolitical safe-haven demand

Q2 net central-bank gold purchases reached 289 tons, up 62% year-on-year

Gold is above the 100-day moving average, with the weekly uptrend channel intact

On the other side:

$4,400-$4,450 has failed to break three times, and profit-taking needs to be absorbed

Core inflation remains above the 2% target, and the Federal Reserve will not dare pivot easily

If the situation in the Strait of Hormuz unexpectedly eases, the safe-haven premium could quickly disappear

Weak retail sales could evolve into recession concerns

Key levels

Upside resistance: $4,450-$4,451 → $4,500-$4,506 → $4,600-$4,800

Downside support: $4,365-$4,370 → $4,320-$4,350 → $4,207

Trading strategy

Short-term traders:

Go long with a small position on a pullback to $4,370-$4,380, set a stop-loss below $4,350, and target $4,420-$4,450. More conservative players should wait for the strong $4,320-$4,350 support zone before entering, with a stop-loss at $4,280-$4,300 and a target of $4,450-$4,500+. If the daily close holds above $4,450, chase the long, with a stop-loss at $4,420 and a target of $4,550-$4,600.

Swing traders:

Once gold holds above $4,400 and breaks through $4,450, hold a core position for a weekly-chart target of $4,600-$4,800, and raise the stop-loss to $4,300.

Long-term believers:

Build positions in batches below $4,300. Central banks around the world are buying—are you? Year-end target: $5,200, betting on the restart of the rate-cut cycle and accelerated dedollarization.

Gold is undergoing a historic repricing—

On one side, the Federal Reserve does not dare raise rates easily; on another, central banks are aggressively stockpiling; and the Strait of Hormuz could ignite at any moment.

The day $4,450 breaks through, you will realize: It was not that gold was incapable—it was that you got off halfway up every time.

What is your cost basis for gold? At $4,400, would you dare chase it? #标普500首次冲破7800点 #我的七夕交易分享 #股票交易分享挑战 $BTC $XAU $XAUT

Dig down along the physical infrastructure and bottlenecks, following the places that are truly holding everything back:
Who controls the core of AI computing?
1. NVIDIA $NVDA
Almost every large model runs on its GPUs. Its moat is so deep that no one can catch up in the short term, and its order backlog already extends several years into the future.
2. Advanced Micro Devices $AMD
The strongest alternative to NVIDIA, with its data center market share continuing to rise. It offers strong cost performance, and major companies use it both as a primary supplier and a backup.
What is most lacking when chips run?
3. Micron Technology $MU
One of the largest suppliers of high-bandwidth memory (HBM). Without it, AI chips are effectively dead in the water. Its products are already sold out for years into the future, with severe shortages.
4. SK Hynix $SKHY
Another global HBM giant. Together with Micron, it has the high-end memory market tightly locked up. AI chip manufacturers are rushing to place orders, but supply cannot meet demand.
What happens when the data cannot move?
5. Applied Optoelectronics $AAOI
Specializes in optical modules and solves the data transfer bottleneck between GPUs. Copper wiring has reached its limits, so optical is the only way forward, and orders are surging.
6. Coherent $COHR
A long-established leader in optical and photonics technology, providing core optical communications components for AI data centers. It has high technical barriers and strong supply capabilities.
7. Lumentum $LITE
An expert in high-speed optical modules and interconnects, focused on turning large-scale AI clusters into truly usable optical superhighways.
Who is spending huge sums to build data centers?
8. Microsoft $MSFT
It is both the largest AI user and the largest buyer. Azure Cloud is directly benefiting from the entire wave, with stable cash flow and an extremely broad moat.
9. Amazon $AMZN
The world's largest cloud services provider, AWS. AI training and inference demand directly translates into revenue, while the company is also developing its own chips to reduce costs.
Where will the power and sites come from?
10. IREN $IREN
A leading example of a Bitcoin mining company transitioning into AI data centers. It controls large amounts of low-cost electricity and existing sites, and is rapidly switching to become an AI computing provider.
11. Nebius $NBIS
It has access to substantial power and data center resources and is purpose-built for high-performance AI computing. It is becoming a new major computing power provider.
How will AI leave the data center and enter the real world?
12. Ondas Holdings $ONDS
Focused on industrial-grade drones and autonomous systems, it is a direct beneficiary of the deployment of physical AI, with orders and partnerships continuing to materialize.
How will space connect AI globally?
13. SpaceX $SPCX
SpaceX has rockets, satellite internet, and the ability to deploy large-scale satellite constellations. Starlink is building a global low-Earth-orbit communications network. If AI eventually expands into drones, robots, autonomous driving, and edge computing, low-latency global connectivity will become important infrastructure.
14. Rocket Lab $RKLB
A dual player in both launch services and satellite platforms, it is a key link for truly sending AI, drones, and remote connectivity into space.
15. AST SpaceMobile $ASTS
It provides space-based broadband directly to phones and drones, enabling AI to stay online in real time anywhere on Earth. The potential is enormous.
Who are their major customers?
16. $OpenAI
One of the world's leading AI model companies and a major buyer of NVIDIA GPUs, HBM, optical modules, and data center capacity.
17. $Anthropic
The AI company behind the Claude series of large models, with rapidly growing AI training and inference needs. As model parameters, context lengths, and agent applications continue to expand, demand for GPUs, HBM, network interconnects, and data center computing capacity continues to rise.
Who else is missing?

After liquidating his positions at the end of 2025 and staying out of the market for six months, top American venture capitalist Peter Thiel has made his move again:

Increased holdings in

1. Amazon $AMZN
largest position, $110 million. The world's largest cloud services and data center operator, it is seeing its electricity demand rise directly as AI computing power surges. The company is also actively developing small modular reactors, positioning it to benefit from the AI energy boom over the long term.
2. Vista Energy $VIST
a major producer in Argentina's Vaca Muerta shale oil region. Its output continues to grow rapidly, benefiting from global energy demand and local policy support, with strong growth potential.
3. Vistra $VST
an independent power producer with extensive nuclear and natural gas generation assets. It is well positioned to provide the stable electricity supply most needed by AI data centers, making it a typical “picks-and-shovels” play.
4. American Electric Power $AEP
a large regulated utility with extensive grid and transmission projects. It can steadily benefit from the growth in electricity demand driven by the expansion of AI and data centers.
5. DTE Energy $DTE
a leading regional electric and natural gas utility. Its business is stable, and grid upgrades directly support the growth in data center and industrial electricity demand, offering both dividends and growth.
6. FirstEnergy $FE
a major power company in the Midwest. Its significant investment in grid modernization positions it to benefit from AI computing's long-term demand for reliable electricity.
7. CMS Energy $CMS
a core electricity and natural gas supplier in Michigan. Its business is stable, and grid upgrades align well with the increase in electricity demand brought by data center expansion.
8. X-Energy $XE
an advanced small modular reactor developer backed by the U.S. Department of Energy and partnering with major companies such as Amazon, positioning itself directly in the next-generation clean and reliable power sector.

Everything he bought is electricity-related!

What I like about @DuskFoundation is the balance between privacy and transparency
Bringing real-world assets on-chain only works if sensitive financial data can stay protected while the network remains compliant.
That’s the direction that makes $DUSK interesting to me.
#dusk

SNDK at $1,735—would you dare chase it?

Look at the surface first: from hell to heaven, all it took was one Investor Day.

SanDisk free-fell from its June ATH of $2,354 to around $1,000, a nearly 60% pullback, as the entire internet shouted, “The storage bubble has burst.” Then came the explosive Q4 earnings report on August 5—revenue of $8.97 billion, net profit of $6.9 billion, and a sharp quarter-over-quarter surge in data center business. Then Investor Day dropped the ace: double-digit revenue growth over the next five years, gross margins heading toward 80%, $15 billion in buyback authorization, and NBM long-term contracts locking in tens of billions in revenue.

The stock rebounded from $1,000 to $1,735, up 73%. Consecutive bullish daily candles broke through the downtrend channel on rising volume, MACD momentum turned positive, and RSI recovered from oversold levels to neutral-to-strong territory.

First: SanDisk is no longer the same cyclical stock.

What was the NAND industry like before? Three years of profits, two years of losses, and one year of jumping off a building. When oversupply arrived, no one could withstand it. The life of a cyclical stock was a roller coaster.

But now things have changed. Management’s NBM has directly locked in approximately 50% of FY2027 shipment volume, as well as roughly two-thirds of FY2028 shipment volume and pricing. The minimum contract value is already in the tens of billions of dollars—the revenue over the next three years is locked in.

Retail investors are still following the old script and panicking that “storage is about to collapse again,” while institutions have repriced it as a growth stock tied to AI infrastructure.

Second: AI storage is underestimated by 99% of people.

Everyone is focused on NVIDIA GPU computing power, but no one is paying attention to one ironclad fact: AI training and inference need not only computing power, but also storage. Inference scenarios have even more extreme requirements for storage throughput and latency than training. SanDisk’s High Bandwidth Flash (HBF) has completed tape-out, with sampling scheduled for 2027, targeting precisely this untapped market.

Third: The perpetual market is betting on confirmation after the U.S. stock market opens.

SNDK perpetual OI has already exceeded $1.73 billion, making it the largest among all stock perpetuals. It climbed from $1,641 to $1,735 over the weekend, a premium of approximately 5.7%, showing that bullish sentiment had been building throughout the weekend. But after the July plunge from $2,354 to $1,000, highly leveraged players are still licking their wounds—will the same script play out after the U.S. stock market opens?

Bulls versus bears—judge for yourself

On one side:

FY2026 revenue of $20.25 billion, up 175% year over year, with net profit of $11.43 billion

NBM locks in tens of billions in revenue over the next three years, significantly reducing its cyclical characteristics

Gross margin of 80%+, operating margin of 75%, and FCF/revenue of 50%

$15 billion in buybacks plus 100% excess cash returned to shareholders

The median analyst price target is above $2,100, with the highest reaching $3,600

From $1,000 to $1,735 in August, with rising volume and trend confirmation

On the other side:

It plunged from $2,354 to $1,000 in July, a 60% pullback that completely wiped out highly leveraged players

$1,700–$1,800 is a heavily contested resistance zone from the previous move, and it failed to break through three times

If AI capex slows or capacity is released, high growth may not be sustainable

Perpetual OI is massive, so the possibility of a false breakout is not low

Macro interest rates remain high, and high-beta stocks are extremely sensitive to negative news

Key levels

Resistance above: $1,800–$1,860 → $1,900+ → $2,350 (ATH)

Support below: $1,700 (heavy premarket activity) → $1,640–$1,600 → $1,500 → $1,270

Trading strategy

Aggressive bulls:

Open a small long position around $1,735, or add on a pullback to $1,700–$1,720. First target: $1,850–$1,900; medium-term target: $2,100+. Stop-loss: exit if it decisively breaks below $1,640–$1,650.

Conservative bulls:

Wait for the U.S. market to open and confirm that it can hold above $1,750 before entering, or wait for a pullback to the $1,600–$1,650 area before getting in.

Bears:

Shorting is not preferred at present. Only consider a short opportunity targeting a retest of $1,500 after a clear, high-volume break below $1,600.

Risk-control rules:

Risk no more than 2% of total capital per trade. This stock’s historical beta is approximately 3.8, even wilder than Bitcoin. Once profits exceed 30%, at least halve the position to lock in gains—don’t get greedy.

It’s not a coin, but it’s wilder than a coin—

Every crypto gambler is looking for a 10x coin in AI, but the real 10x opportunity backed by hard logic is in the perpetuals of this “traditional stock.”

The day $1,800 breaks through, you’ll realize that storage is the most reliable pick-and-shovel play in AI, and SanDisk is the one holding the most shovels.

What is your SNDK cost basis? After the U.S. stock market opens, are you betting on a breakout or a pullback? #标普500首次冲破7800点 #我的七夕交易分享 #股票交易分享挑战 $SNDK $MU $SKHY

$AEON : Could it be the payment layer of the AI economy? 👀🔥
$AEON @AEON_Community broke out of the $0.05–$0.07 accumulation zone and strengthened its momentum as it moved toward the $0.08+ region.
But the real story is not the price. It is AI agents starting to make payments.
🤖 AI → Agents
⚡ Agents → Autonomous Transactions
💳 Autonomous Transactions → Agentic Payments
🔗 Payments → Settlement
As the x402 narrative gains momentum, machine-to-machine payments are becoming increasingly important.
AI agents will not just chat; they could become economic actors that independently pay for APIs, data, computing power, software, and various services.
This is where $AEON stands out:
🤖 AI Agents
💳 Crypto Payments
⚡ Settlement Infrastructure
If the use cases for AI agents expand and agentic payments become real economic infrastructure, the payment and settlement layer this sector will need could create a significant opportunity.
That is why the thesis for $AEON is not merely:
“An AI token is rising.”
The bigger thesis is:
AI → Agents → Autonomous Transactions → Payments → Settlement → $AEON
Price movement could be just the beginning of this thesis. 👀🚀
#AEON #AgenticEconomy #AIPayment