Orbit Post Sitemap
#财报观察员: AI infrastructure financial reports debut in succession Recently, global AI infrastructure companies have been intensively disclosed in their financial reports, with chip, server, storage, and cloud vendors delivering their report cards one after another. The AI industry has officially entered its "accounting moment," and the market is no longer just chasing capital expenditure stories, but is placing greater emphasis on order conversion, cash flow, and profit quality. Upstream hardware performance was the most prominent. Storage, AI servers, and high-speed interconnect hardware companies benefited from expanded computing power, with revenue and profits rising sharply. NVIDIA is about to enter a critical earnings window, with the implementation of a $500 billion computing power financing framework. The market is closely watching GPU shipment guidance and actual downstream procurement demand. Although storage companies like SK Hynix hit record profits, their stock prices fluctuated sharply after the financial report was released, reflecting market concerns about whether the high prosperity could be sustained. At the cloud vendor level, Microsoft, Amazon, and Google have continuously raised AI capital expenditures and invested heavily in computing infrastructure, but massive capital investments keep suppressing free cash flow, creating a pattern where "hardware profits, cloud providers expand revenue." On one hand, orders and revenue are growing rapidly; on the other, data center construction and GPU procurement bring massive capital expenditures. When ROI will improve has become a core question for institutions. Current market divergence is intensifying: an optimistic view believes AI computing power demand will remain rigid for the long term; The risk points are concentrated in the rapid depreciation of hardware iterations and the false demand driven by leverage. If downstream AI commercialization falls short of expectations, the high capital expenditure model will face pressure. Earnings season is currently filtering for stocks with truly fundamental strengths, and thematic speculation is gradually fadingThis afternoon, while writing a weekly report at a café, two people at the next table were chatting loudly. One was talking about an AI coin he bought last month, which doubled but didn't come out, and now it's fallen back to cost. The other said, "At least you didn't lose money," then fell silent. I glanced down at my screen. $BTC's price had moved less than 300 dollars from morning until now, spending the whole day digesting last week's high-volume bearish candle.
ETH gas prices have hovered in single digits for a week. Based on past experience, this ultra-low gas won't last long, as on-chain activity will eventually revive. But this time is different: cross-chain bridge traffic on Solana is steadily rising, with daily funds bridged from ETH doubling compared to last month. ETH is losing some of its original liquidity.
I noticed an address on $AAVE performed a very complex operation today. First, borrow USDT with WBTC, then exchange USDT for USDC, then use USDC as collateral to borrow ETH, and finally exchange ETH for stETH and deposit it into the liquidity pool. This entire process involves three lending protocols and two DEXs, with an annualized return of about 0.8%. Those who use this strategy are clearly not chasing high returns but seeking positive skew return opportunities, earning from price differences that others ignore.
$LINK today, some addresses continued to accumulate. One address has been withdrawing from exchanges continuously over the past week. Each withdrawal amount is small but the frequency is extremely high, averaging one every two hours, with cumulative withdrawals exceeding 100,000 LINK. This kind of high-frequency small amount is more like an oracle node supplementing collateral, not a retail activity. $MKR Others are doing similar things. One address withdrew a considerable amount of MKR from Coinbase today and directly deposited it into Spark's sDAI pool, living off savings interest while continuing to hold MKR's governance exposure.
$UNI The front-end long-tail trading volume shrank slightly today, but there was a detail—a market maker address simultaneously adjusted its position on both $CRV and $CVX tokens, swapping CRV for CVX, with the amount exactly matching one to one. This usually means some expect the value allocation within the Curve ecosystem to change, and CVX is undervalued relative to CRV. $PENDLE's yield curve steepened today, with some continuously buying long-term PT. Considering last week's large $ENA position, some are planning a multi-month interest rate trading strategy.
$SOL Today, several cross-chain stablecoin inflows came from the same ETH address, with more shares split than yesterday. The smallest was only a few hundred U, and the largest just over ten thousand. This approach is like spreading deposits across multiple lending protocols to earn interest, while maintaining liquidity that can be withdrawn at any time. If you see more and more addresses doing the same, it shows the market truly lacks better profit opportunities. $SUI and $APT didn't hit new lows today, but their rebound was weak and basically moved slightly with the market. $SEI's trading volume was even lower than yesterday, and the thickest buy order on the order book was only over $30,000—a depth that could break even a slightly larger retail investor. $TIA and $INJ didn't have independent moves today, completely following the market's rhythm, with volatility narrowing to a recent low.
The AI sector did not continue yesterday's relative strength today. $FET and $RNDR briefly surged in the morning session but were quickly suppressed. The selling orders came from several fixed addresses, not the same entity, but with very regular intervals, placing a batch of sell orders every fifteen minutes, as if executing a TTAP selling strategy. $TAO Today, there was a large transfer from the exchange to a private wallet, nearly six figures, a volume of withdrawals usually not something retail investors can make. $AR continued to shrink and move sideways, $NEAR there was an address continuously withdrawing from Binance, withdrawing at fixed times every day for five consecutive days, accumulating quite a lot.
The RWA line quieted down a bit today, but $ONDO and $OMNI address activity hasn't completely stopped. The address that tested slippage range yesterday made no new moves today, but another new address crossed from Polygon to ETH and then bought US Treasury tokens—not a large amount, the path is exactly the same—indicating someone is running a structured automated strategy. A large USDC to DAI swap occurred in $CRV's stablecoin exchange pool today, with slippage well controlled, and the execution price was almost equal to the market price. This kind of transaction looks like an institution adjusting stablecoin allocation, not arbitrage.
The Meme sector was collectively quiet today. Volatility for $PEPE and $DOGE has dropped to recent lows, with bid-ask spreads widening compared to usual, and market makers are reluctant to provide much liquidity at this level. Trading volumes for $BONK and $WIF have also shrunk sharply. Occasionally, someone in the group shouts, "Is this bottom?" and the comments below are either memes or "If it drops half more, I'll buy." On the $FLOKI side, there was an address change today—a small transfer from the exchange to the contract address. This kind of operation is usually someone preparing to deploy new liquidity in Memes, but the amount is...
Net buying is obvious: $LINK, $AAVE, $MKR, $ENS, $LDO, $RNDR, $FET, $TAO, $MNT, $METIS. Continuous net selling: $ARB, $OP, $STRK, $BLAST, $SUI, $APT, $SEI, $TIA, $INJ, $PEPE, $BONK, $WIF, $FLOKI. These data only show today's capital flow; waking up tomorrow may look like a completely different situation.
In the evening, I went downstairs to buy a cup of coffee and took a couple of sips at the entrance. The milk tea shop across the street had a queue stretching onto the sidewalk, where a few young people were discussing where to go for the weekend. Most things in daily life have nothing to do with the crypto market. When the market is quiet, you can actually see who is actually working and who is just trying to ride the wave.
The cup was empty, the sky was dark, and it was time to go back and check the data. #财报观察员: AI infrastructure earnings report debuts one after another
Damn! The two-year AI infrastructure frenzy has come to an end, and the market is finally going to conduct an autopsy: after burning hundreds of billions, have orders really turned into profit?
No matter how good the orders on the PPT look, you still need to provide real gross margins and cash flow; otherwise, the market will just be thrown away like garbage.
In the optical module segment, Lumentum is the bottleneck. Both Google TPU and NVIDIA GPU rely on its laser and optical components.
Q3 revenue has already reached over $800 million, nearly doubling year-on-year, with operating profit margin pushing to over 30%. Wall Street expects next quarter to aim for $1 billion, doubling again.
After tonight's post-market report, if revenue and profit margins exceed expectations again, the short-selling logic still focused on storage will become even tougher; If the market drops, the optical rally will immediately return to its original state.
A well-known trader on X also bluntly said: AI clusters have expanded from thousands to hundreds of thousands, computing power is no longer the only bottleneck, and the real problem is that data cannot be moved. Bandwidth scarcity is replacing computing power scarcity, and Lumentum is stuck in the middle. Some bluntly say its capacity is already scheduled through 2028, and this is not a cycle, but a structural shortage.
SpaceX is in another extreme case: after the first batch of over 900 million shares was unlocked, the stock price actually jumped from 108 on the day of the unlock to 136, a 26% increase. For now, the market is still willing to pay for the long-term story of AI aerospace infrastructure.
But on August 20, the second wave of 319 million shares will be released, with even larger volumes in September and October. Early holders have been sitting on paper for years, and now they can finally cash out.
A KOL on X thinks: Being able to sell doesn't mean you have to dump, but supply is real. Whether the stock price can hold up depends on how the market prices the Terafab project. If it can't hold, bears will just count the money; If it can, the narrative can hold up for a few more seasons.
The remaining financial reports this week are meant to be used for autopsies. CoreWeave's revenue continues to double, with orders piling up to nearly $100 billion, yet GAAP is still losing huge amounts, and capital expenditures are burning through like a life.
Cisco directly raised its AI infrastructure order expectations from 5 billion to 9 billion. Colleant's orders are already scheduled through 2028. Once Applied Materials is released, it becomes clear whether chip factories are truly expanding production or just boasting.
In fact, everyone understands that revenue growth has become a ticket; what truly determines the stock price are the backlog order conversion rate, the actual number of megawatts online, whether interest costs are out of control, and whether the pace of cash burn can be reduced.
Some even call CoreWeave the cleanest touchstone for whether AI infrastructure is a bubble. The demand looks strong, but whether execution keeps up is the real problem.
The track is rapidly diverging. Some companies have already started to truly make money, with gross margins rising; Others are still frantically grabbing land, raising funds, and burning cash, hoping their utilization rate will climb up before the music stops.
The storage side just delivered explosive financial reports, but still got smashed. Making money now isn't enough; you have to prove how long you can keep making it. The same goes for those burning money for growth; the market has started calculating returns for every dollar.
The story ended, and the market began to dig up old grievances. Whoever could turn paper orders into higher gross margins and cash pockets would continue to be honored. The rest were all cannon fodder for the next wave of buyers. #Strategy再卖1690枚BTC, corporate financial pools are diverging
Sigh, Strategy sold again.
From August 3rd to 9th, 1,690 BTC were sold at an average price of $64,262, cashing out 108.6 million. All of it was used to buy back STRC preferred shares. Including the 1,638 BTC sold the previous week, a total of 6,916 BTC were sold over six weeks, recovering 429 million. The position dropped to 840,447 BTC.
But this is just one hand turning the other. In the same week, Strategy sold 6.5857 million shares of MSTR common stock through its ATM program, raising a net raise of 653 million, of which 650 million was directly transferred to dollar reserves. Dollar reserves grew from 4 billion to 4.65 billion.
These details are worth pondering.
First, selling even at a loss. The average holding price was $75,385, and this time selling at $64,262 means a single loss of 18.8 million. But STRC preferred shares have already been discounted to $96, so you have to buy back.
Second, cash reserves are the real bottom line. US dollar reserves are 4.65 billion, with 785 million remaining. Saylor admitted through his actions: having no money in your books is even scarier than a drop in Bitcoin.
Third, BitMine took the completely opposite path from Strategy. Strategy was selling BTC to hoard USD, while BitMine bought 7,391 ETH at an average price of $1,888. One was contracting, the other expanding.
Selling 429 million in six weeks is not large for a volume of 840,447 coins. But this is no longer a question of "sell or not," but "when will selling stop?" When the biggest bulls start selling coins to prove they can pay, the market believes not in how much you sold, but on how much longer you can hold on $BTC #本周三CPI公布, will the pricing for a rate hike in September be rewritten?
#霍尔木兹海峡通航协议未落地, oil price risks are heating up
Today, the main characters influencing $BTC emotions,
Maybe it's not a certain chain, nor is it a new token,
But a barrel of oil.
On August 10, tensions around the Strait of Hormuz flared up again, with oil prices rising about 5%; During the same period, the S&P 500 index slipped slightly by 0.06%, and the Nasdaq dropped by 0.32%. The U.S. Bureau of Labor Statistics will also release the July CPI at 20:30 Beijing time on August 12.
Rising oil prices will reignite market concerns about inflation, and inflation expectations will affect interest rates and risk asset valuations. BTC has recently hovered near $65,000 without a clear statement, behind not only internal bullish and bearish battles in the crypto world but also anticipation before macro events.
Sometimes, understanding a BTC K-line,
We really need to see what happened with crude oil first.$NBIS Options have divergence
Spot short-term weakness and delta liquidity flows weakened, but options did not panic; instead, they gave calls a premium.
The price closed at $184.11, down 2.05%, with the night session around $186.
190 is a key position, with multiple different market structures concentrated here. 190–195 was originally one of the most active options areas.
After the spot market broke through, it was impossible to accept the high price and gave back everything, indicating that the chasing funds failed to control the market the day before the earnings report.
184 happens to be the largest trading area in the dark pool, effectively returning to the chip exchange zone.
The implied fluctuation in the earnings is 11.01%, at a price of 184, roughly within the range of 164–204.
OI is Barbell type, with 150 puts, 190, 250, and 320 calls expanding, betting on market volatility.
If the price before the earnings report is above 190, then overall it is on the bullish side.$BEAT Is the rebound weak? Is it like $LAB $RAVE where the big players are running away? Will it keep falling?
1. Excessive previous gains
* BEAT has seen several times its rise, with heavy profit-taking in the market.
* Once funds withdraw from these projects, they often fall faster than they rise.
2. Recently, it has consistently underperformed the market
* In the past week, BEAT was one of the underperforming tokens, while some altcoins have already started to rebound during the same period.
From the perspective of "controlling the market."
If any of the following situations occur, I will be more alert to the dealer's withdrawal:
* The top 10 addresses have a very high proportion of open positions;
* Project wallet continuously transferring coins to exchanges;
* Community operations suspended;
* Market makers withdraw their bids;
* Trading volume is declining, but prices continue to decline.
Currently, the publicly available information is more likely:
Unlocking + profit-taking positions fleeing + market makers weakening support
It may not necessarily be a run, but it may have already entered the late stage of dealer distribution
Key points:
(1) Top 10 Holdings: 84.34%
This is the most alarming data.
Regular items:
* Top 10 General: 20%-50%
Moderate Control:
* Top10 50%-70%
Height control:
* Top10 ＞80%
And here at BEAT:
Top 10 holdings = 84.34%
This indicates that the chips are extremely concentrated.
But here's a problem:
If most of this 84% is:
* CEX exchange wallets
* Liquidity pools
* Project vault
Then the risk is not as high.
If 10 private addresses control 84%, that's a different story.
(2) Developer Holdings <0.01%
The data actually doesn't look like a runaway market.
Because many rug projects will see:
* Developer holds 10%
* 20%
* 30%
Then they directly dumped the price.
Developers have cleared out their stocks
This means:
It might be:
* Tokens have already been allocated to market makers
* Transfer to Foundation Wallet
* Transferring to institutional wallets does not necessarily mean malicious abscondence.
(3) Smart money adds to the position
Screenshot shows: Smart money adds to the position
Although this label is not necessarily 100% accurate.
But if there are indeed wallets with high profit rates on the chain that keep buying recently,
Explanation:
There is still capital in the market willing to buy in.
This doesn't quite match the bookmaker's complete abscondence.
(4) Price drops by 51%
This actually fits:
After the dealer pays, the market falls in shadow
Instead of:
The market makers flee and the price plunges
Differences:
Runaway:
* Drops 80%-99% in one day
* Liquidity disappears
* No one is taking over
Distribution:
* Falling for several consecutive weeks
* Every rebound gets smashed
* Trading volume gradually decreased
BEAT is more like the second type.
My judgment
Probability estimation:
* Banker has already absconded: 20%
* Dealer still in the distribution phase: 50%
* Banker Shakeout Preparation for Second Wave: 30%
Currently, the most dangerous signal is not developer liquidation.
Instead:
Top 10 holdings: 84.34%
This indicates that BEAT's future price movements will largely depend on a few major players.Japan's national debt is collapsing
The 2-year yield reached 1.615%, the highest since March 1995. The 5-year yield was 2.09%, a 31-year high. The 10-year yield was approaching 2.805%, just a breath short of the 3% warning line
An economy with debt-to-GDP ratio over 240% and interest rates soaring—this scene is too beautiful to look at.
What is the market afraid of?
Expectations for rate hikes are growing more urgent—the probability of a rate hike in September has soared to 66%, compared to 30% at the end of last month. The 2-year yield has soared, betting that the Bank of Japan will take action
What's even more frustrating is that the government is pursuing expansionary fiscal policy, the central bank wants to raise interest rates, and both sides are working against each other. Government bond auctions have even suffered a crushing demand collapse, directly pushing yields higher
The bigger problem is—
Japan is the world's largest holder of U.S. Treasuries. If forced to sell U.S. Treasuries to save its bond market, U.S. Treasury yields will be pushed higher, causing global asset pricing anchors to shake accordingly
For the crypto world, once expectations of liquidity tightening form, risk assets suffer
I'm keeping an eye on Japanese bonds. If the 3% rate breaks, it might not be Japan's problem, but a global oneThe most valuable thing on the board isn't the move where the king gets checked, but the unmoved rear that the opponent has already mobilized in the shadows. Sandisk's better-than-expected earnings and EPS from the previous quarter planted a less-than-consensus hint at the midpoint of next quarter's guidance—this contrast isn't a beginner's miscalculation, but a double threat commonly used by grandmasters: on the surface, you get a piece, but in reality, your rear wing is already wide open. The market's violent swings after the earnings report are like a heavy cart without the protection of the pawn chain—clearly there is a straight line to go, but you don't realize the opponent's bishop has already diagonally blocked all return defense routes.
The truly valuable chess is never in the move you already see, but in the move you haven't yet seen. Investor Day, August 13th, is the moment both sides agree to review the mid-game game. Weak guidance—is it a temporary plug or an early surrender to weak real demand? The supply and demand of flash memory chips determines whether the mid-game pawn structure is stable or fragmented; The AI storage roadmap determines whether you still have the ability to launch a king-wing all-out attack. A $14 billion buyback is exchanging barges for king-wing soldiers—you lose immediate offensive power but gain a theoretical win rate in the endgame. But having one more pawn in the endgame is not victory; it is the beginning of torment. The timing of exchange is always more important than quantity; if you trade at the wrong time, it's like returning victory to the opponent.
I've seen too many players scramble when their opponents discard a piece. A true grandmaster will ask himself at the moment the opponent discards a piece: Which line does he want? The true value of this guidance low may not be telling you weak demand, but rather that management has already anticipated the next mistake in market consensus. When everyone is betting on next quarter's performance, smart players are already calculating the slight difference in the opponent's king's position in the endgame. XIBM's synergy is like two shadow chess pieces on a board—same opening, different variables. You think you're seeing the same game, but in reality, the timers on both sides are already separated by more than a beat.
In the middle game, the scariest trap is not the opponent making a strong move, but the move that seems smooth but quietly shifts the center of gravity of the entire board. This guiding midpoint is exactly such a gentle strike. It doesn't cause the market to collapse, but it forces everyone to reassess their relationship with the chessboard. Investor Day is more like a "halftime lock"—each side has time to test the errors of their assumptions. But only after the lockdown ends, the real situation emerges.
The market wants to find answers on investor day. But the master's answer is never at the review meeting. The review meeting is just something that makes your opponent believe you're half-convinced. So, don't rush to judge that weak guide as a defensive flaw. Discarding a piece may be the only way to a decisive win. When the midpoint of that guide written on paper is like a toppled pawn, falling into a gray consensus zone—the real question is never why it falls, but what appears on the board after it falls.
You know that wasn't Wang Qian's fall, but Wang Yi's abyss #sandiskinvestordayThe bullish whale suddenly turned around, with $8.05M short positions pressing on SKHX
A familiar face on the 7-day PnL leaderboard, 0x66f4... 8836, a swing address with $20.36M in equity and $1.75M in historical profit, usually prefers to side with the bulls.
But in the five minutes that started at 5:33 a.m., he opened 2,000 short positions, executing 7,950 SKHX contracts worth $8.05M, with the average price around 1,012.2.
This is completely opposite to the usual directional sense in his 58 historical trades, like a sudden shift in stance.
Currently, the account is cross-marginalized, with no other positions in the same direction. This trade is almost a gamble.
Public data shows a win rate of 33.3%, which isn't high, but the PnL curve holds up, suggesting that stop-loss or profit-loss ratio management might be effective.
If SKHX cannot quickly reclaim the cost zone, whether they will continue to shorten or reverse will be key to observing their true intentions.
If you like my sharing, please give me a follow$JD Stock price formed a narrow consolidation range at $33.47, with a mismatch between $45.2 billion in market cap and $1.32 trillion in annualized revenue, forming the core battle point for whether gross margin can support valuation recovery.
From a price structure perspective, $JD's daily trading volume remained at $7.65 million, with the price in a sideways consolidation state. Compared to peer stocks, BABA's circulating market capitalization is $317.15 billion, and PDD's circulating market cap is $132.38 billion. Their $45.2 billion market cap shows a valuation center discount characteristic.
In ranking the driving factors, the top priority is whether gross margin matches the intensity of capital expenditure, followed by order growth rate and changes in market share. The suppression of the overall valuation center by macro interest rates is considered an external variable.
In the valuation recovery scenario, if order growth recovers and quarterly gross margin stabilizes, supply chain monetization efficiency is verified, and the price-to-sales ratio has 20% to 50% expansion potential, prices will break out of the consolidation range near $33.47 and seek an upward breakout.
In the downside test scenario, if the price war intensifies again to squeeze profit margins and AI capital spending falls short of expectations, valuation pressure will trigger sell-offs, and market cap may face a downside test approaching the $45.2 billion halving level.
After the downward trend is confirmed, if subsequent company buyback amounts significantly increase, the downward rhythm will be interrupted and converted into accumulated accumulation within the range.
The failure of this simulation lies in the rise in macro interest rates. If the external interest rate environment pushes up and suppresses the overall valuation of Chinese concept e-commerce, the logic of upward growth brought by stabilized internal gross margins will immediately fail.
Over the next 7 days, focus on tracking the implementation of gross margin and revenue growth in quarterly data, as well as institutional position changes and order backlog data disclosed in the 13F report.
#三星钱包将接入稳定币, payment scenarios continue to expand. #霍尔木兹海峡通航协议未落地, oil price risks are risingLife within the system is really tough now, with many people overdue their loans.
The president talked about finance, saying that many of his overdue debts are now within the system, and this data is growing rapidly.
I said, aren't the provident fund customers your high-quality clients?
He said yes, because of the high quality, they used to be very wide.
He said, 'Do you know what they do? Most are speculators.'
Secondly, they are involved in investment; many projects are backed by their own investments,
Because they have circles and resources, and have mastered many excellent projects,
They just vote themselves and find someone to hold on their behalf. Who knows,
Good projects were good before, but now, in this situation,
There are no truly good projects left; before, it was easy for them to get loans,
Many people love to fuss, and now things are getting tough.
Some have problems that need money to settle them, generally three types of problems,
Women's issues, disputes with businessmen, and dirty butts—he used to be able to suppress them, but now it's clear they can't.
So the last wave of spoilers will hit this group,
Ordinary people would have died two years ago, but those with better conditions could endure for a long time.
Actually, what was the previous period of internal government dividends? It wasn't that wages within the system were high,
Not high. Even if you're on the coast, it's not high compared to the incomes of local wealthy small business owners.
But why do they have money, or rather, get rich?
It's not really about corruption; corruption is only qualified for top leaders or those in positions with real power.
Others can't get much money, and the boss won't give money to those who don't have a decisive impact.
They find it easy to get credit, easy to speculate, easy to get information, familiar with policies, and then start a side business with projects.
They weren't here for salaries or corruption. Which old comrade doesn't own three or five houses?
Anyone who used to work at a state-run hotel could own three apartments—two old, run-down and one new, improved.
Now it's no longer possible; it's hard to make money霍尔木兹海峡的海床在颤抖，这不是地质运动，是地缘承重墙在发出应力断裂的呻吟。我盯着那组油价数据——WTI站上82.13，布伦特摸到87.72——就像看着一栋超高层建筑在台风里记录摆幅，风洞试验报告显示阻尼器已经满载，但塔吊仍在旋转。
谈判桌上的“过境费”没有敲定，这在我眼里等同于桩基检测报告缺席。任何结构工程师都清楚，摩擦桩和端承桩的承载力计算方式完全不同，而伊朗和阿曼连摩擦系数都没确认，这座名为“稳定通航”的桥梁就敢把施工缝留在霍尔木兹最窄的航道上。特朗普把琼斯法案豁免又延长了90天，那不过是往混凝土里掺了早强剂——表面上3天即可拆模，但水化热的累积足以让大体积混凝土内部开裂。
真正要命的不是油价今天的跳空，而是“长期可展性”这个建筑学核心命题。制裁框架是外围幕墙，保险条款是防火分区，航运规则是疏散通道——三者如果不能像CBD的钢结构节点那样咬合严密，那么任何一艘LNG船穿越海峡时，都是在做一场没有冗余度的荷载试验。我见过太多白皮书上的摩天楼，效果图能直插云霄，却在打桩阶段就碰到了溶洞。
市场对霍尔木兹的风险溢价，本质上是给“竣工日期”这个不可信承诺的贴水。油轮改道好望角，就像把核心筒偏移三度，大厦的偏心率会成倍放大。而XQQQ这类资产在恐慌指数触达FearAndGreedIndex的贪婪区时还保持联动，这本身就说明资金在把石油美元和科技股的“垂直交通”焊在了一起——电梯井里如果进了海水，顶层复式卖得再贵也只是灾难片的布景。
现在各方都在声称自己能管控风险，但建筑史从不承认嘴上的强度等级。霍尔木兹海峡的“活荷载”每天都在随油轮吃水线变化，而“结构健康监测”系统至今没有一份三方签字的验收单。我可以画出最漂亮的悬索桥图纸，却无法说服自己踏上一座没有做过风洞试验桥面板的桥。
结构失效从来不是瞬间发生的，它只是最后才被肉眼看见。 #hormuzdealunresolved📊 $CL合约爆仓速递（8月16日）
根据爆仓数据，这波空头被狗庄按在地上疯狂摩擦了。。。
时间 总爆仓 多单爆仓 空单爆仓
1小时 $5,888.01 $87.09 $5,800.92
4小时 $3.42万 $3,522.65 $3.07万
12小时 $15.35万 $1.51万 $13.83万
24小时 $168.65万 $8.53万 $160.12万
从$CL爆仓数据看，1小时空头爆仓碾压多头，空头爆仓量是多头的66倍，逼空闪击开局即核爆级烈度；4小时空头优势持续，比例约8.7倍，逼空全面爆发；12小时空头仍遥遥领先，比例约9.1倍，逼空贯穿短中周期；24小时空头爆仓飙升至160.12万美元，是多头的18.8倍，狗庄在CL上完成了对空头的全周期屠杀——短中长周期空头被全方位定向爆破，多头仅有的反抗杯水车薪，累计爆仓突破168万美元。空头血流成河，逼空行情势如破竹。大家控制好仓位，别被来回收割。
🔥 市场风向标 | 8月16日
今日三条热点，指向同一主题：AI基建的资本盛宴，正在迎来一场从"烧钱"到"赚钱"的残酷大考。
🏗️ AI基建财报接力：市场只认"真金白银"
Q2财报季，华尔街的逻辑已彻底转向。亚马逊AWS营收同比猛增37%，微软Azure飙升43%，三大云业务合计增长达48%。真正推动亚马逊跻身3万亿美元市值俱乐部的，正是AWS创下18个季度以来新高的增速。
然而，市场并非"见AI就买"。Meta交出超预期财报，却因AI投入尚未形成独立收入来源而盘后下跌；英伟达全周仅微涨2%，投资者正屏息等待8月26日财报。市场奖励的已不再是"谁投得多"，而是"谁赚得快"。
📊 CPI今夜公布：9月加息的天平悬在半空
北京时间8月12日20:30，美国7月CPI将公布。克利夫兰联储预测7月整体CPI环比仅微涨0.09%，但核心CPI预计环比上涨0.21%，较6月的环比持平有所反弹。
当前市场定价9月加息概率约44%-55%。若今夜数据超预期，鹰派阵营将迅速扩大；若温和，加息预期则可能进一步消退。这份数据，将是决定9月FOMC会议走向的第一块多米诺骨牌。
💰 英伟达推动5000亿AI基建融资：GPU变成"可投资资产"
8月10日，英伟达宣布与阿波罗、贝莱德、黑石、高盛、KKR六大巨头合作，设立独立算力融资平台，目标撬动超5000亿美元第三方资本。黄仁勋亲赴华尔街，六家机构无一缺席。
然而消息公布当天，英伟达股价反而下跌约2.8%，市值蒸发超700亿美元。《大空头》原型迈克尔·伯里公开警示，"循环融资"模式可能重演2000年互联网泡沫破裂前的借贷乱象。黄仁勋则强调，AI算力已是"等同于电力、互联网的新时代刚需基础设施"。
💎 总结
AI基建的财报季证明了一件事：市场不再为"故事"买单，只为"回报"定价。英伟达的5000亿融资计划，是这场资本游戏的最高潮——也是最大的赌注。而今夜CPI的每一个基点，都可能决定这场赌局的宏观底色。#财报观察员：AI基建财报接力登场
#本周三CPI公布，9月加息定价会改写吗？
#英伟达推动5000亿美元AI基建融资 Never would have expected
Shorting $SPCX was taught by Lao Ma
Indeed, there is no such thing as a certain opportunity in this world
Everyone thought the unlock would crash
The final result is a pull-up
The market always goes against what most people expect
However, I am still not optimistic about the $SPCX market this year
The unlock on August 6 is just the appetizer
Up until December this year, there are stepwise unlocks
Old Ma wouldn't accept everything without hesitation
Selling pressure still exists. #财报观察员: AI infrastructure earnings report debuts in succession #英伟达推动5000亿美元AI基建融资 #本周三CPI公布, will the September rate hike pricing be rewritten? 中概电商赛道的估值分化在收窄，支撑供应链资产的买盘与价格战带来的盈利隐忧在特定价位交错。
$JD 股价徘徊在 33.47 美元附近，单日成交额维持在 765 万美元，整体呈现窄幅整理格局。
在年化收入达到 1.32 万亿美元的背景下，其 452 亿美元的流通市值明显低于同行阿里与拼多的市值体量。
体量与估值之间的断层能否转化为修复动力仍待确认，核心在于毛利率是否匹配资本开支。
若订单增速恢复且毛利率企稳，市销率存在 20% 至 50% 的扩张空间；一旦宏观利率上行压制整体估值中枢，这一上行逻辑将宣告失效。
若价格战加剧挤压利润且 AI 资本开支投入不及预期，市值存在向 452 亿美元减半位置靠拢的风险；若后续回购金额显著放大，下行节奏将被中断。
机构持仓在最新报告期的变动与订单积压状况，将直接证伪或确认当前估值探底的判断。
未来几天最值得持续跟踪的变量，是毛利率与营收增速在季度数据中的实际落地表现。
#标普收盘再创新高，8000点预期升温 #CLARITY表决推迟至9月，监管窗口后移 #伯克希尔结束净卖出，重启大额配置链上新开大仓：SKHX遭遇巨鲸级账户$4.54M空单拆压
SKHX盘口刚刚被连续拆单砸空，一笔$4.54M的新空单正在快速成交。
下单地址0x66f4...8836是7日PnL榜和巨鲸权益榜上的常客，账户权益$20.36M，历史盈利$1.75M，胜率33.3%，58笔交易里大多偏多头波段。
这次却在SKHX上开了4.49K张空单，均价1012.36，用1119笔碎单完成，不是老仓浮盈加码，是纯新仓。
一个长期偏多的巨鲸突然转向，本身就值得多看两眼。目前该账户没有同向持仓，如果接下来继续补空，说明对这段行情的看法已经变了。
如果喜欢我的分享，麻烦点个关注#现货ETF资金分化, BTC selling pressure remains. Last week, $BTC and $ETH spot ETFs combined saw net inflows of about $1.1 billion. Institutional capital inflows were originally a good signal, but the latest data shows divergence: BTC ETFs have returned to net outflows, while ETH continues to see modest inflows.
Meanwhile, on-chain whales and miners are still moving out of BTC. So I think what really deserves attention in the short term is not "whether ETFs are flowing in or out today," but rather:
Can new buying pressure continue to absorb the selling pressure from old chips? ETFs have indeed changed BTC's capital structure, but they cannot eliminate cycles. Miners, early whales, and long-term holders cashing at high levels, as long as selling pressure persists, new capital will keep passing through.
ETH's current relative strength is also worth watching. If ETFs continue to see net inflows while BTC funds keep outflowing, short-term capital may show signs of rotation from BTC to ETH.
But if I had to choose between two now, I would still value $BTC more
The reason is simple: before the macro environment is fully clear, BTC remains the core asset for institutional funds entering crypto.
Next, I will focus on three things: whether ETFs can continue net inflows again, whether whale selling pressure has eased, and whether BTC spot trading volume can expand.
If all three improve at once, I will believe the next round of rally will truly have a capital base. Ultimately, the market isn't about who shouts louder, but whether the market has enough money to push the sell.Uh...... Retail investor subscriptions for Unitree's IPO exceeded 8,000 times.
Is there absolutely huge demand for purely humanoid robot companies?
I am a member of the U.S. Agile Robotics Club (SoftBank, $NVDA, $AMZN, Foxconn, etc.), and achieved a pre-financing valuation of $2.5 billion through $CCXI.
My view is that Unitree's IPO valuation could exceed $30 billion (from pre-IPO investors), which could attract more attention from leading U.S. investors within a week or two.
In any case, this is simply an absurd request.
#财报观察员: AI infrastructure earnings report debuts in succession. #本周三CPI公布, will the September rate hike pricing be rewritten? #英伟达推动5000亿美元AI基建融资 Grayscale submitted Form RW to the SEC, officially withdrawing its applications for Cardano, Hedera, and Polkadot ETFs. No shares were issued, so the ETF was directly canceled.
After the news broke, ADA, HBAR, and DOT fell more than 2% in 24 hours. HBAR fared even worse, having dropped 30% in the past two months and now at about $0.068. Market expectations for a "counterfeit ETF channel" are cooling down.
This is not because Grayscale is pessimistic about knockoffs, but rather, under the triple pressures of regulatory windows, liquidity, and institutional demand, preserving the mainstream (BTC/ETH) is the key priority.
Do you still hold $ADA, $HBAR, $DOT? Will this withdrawal affect your position decisions?
#现货ETF资金分化, BTC selling pressure remains #本周三CPI公布, will the pricing for a rate hike in September be rewritten? MicroStrategy has been selling coins recently. I see some friends worried it might crash, leading to a $BTC crash
Actually, there's nothing to worry about:
1. WeStrategy selling coins is normal; it's an investment institution whose profit model is buying low, selling high. The initial promotion of "never sell" was just to rally more people to join the sale, not to be fully trusted. Even if it doesn't sell in one cycle, it will sell in the next.
Moreover, it had already sold in the previous cycle a few years ago, so this isn't the first time selling coins.
2. During this cycle, MicroStrategy will not crash, because it won't be held back by the annual interest of 1.76 billion.
In fact, MicroStrategy has many methods. First, issue additional common shares. Its common shares have an additional issuance space of about 24.6 billion. Second, issue additional perpetual preferred shares. Its perpetual preferred shares have an additional issuance potential of 25 billion. Third, sell $BTC, with a total BTC value of 58.5 billion.
Moreover, when it really doesn't care about goodwill, the annual interest it has to pay is only 2 trillion yuan. No other interest payments are illegal, and their contract terms are highly flexible.
3. Even if MicroStrategy wants to sell coins to repay debts, BTC's current market cap is 1.2 trillion, the daily spot trading volume is 30 billion, and the average daily trading volume of futures and derivatives is 150 billion. With such a large trading volume, it won't be dragged down by the annual 1.76 billion in selling.
#Strategy再卖1690枚BTC, corporate financial pools are diverging Too many $BTC
The main reason is that CPI data should be acceptable, at least not exceeding expectations.
As for oil prices, although July saw an increase, the average price was not much higher than in June.
Regarding wages, last week's wage data showed both annual and monthly rates below previous values, especially the monthly rate, which is significantly lower. Wages are an important component of goods and service costs.
On the consumption side, the preliminary consumer confidence index for July was lower than the previous value, indicating no trend of demand growth pushing prices higher.
Regarding the 'pubic hair' theory, we previously discussed that nonfarm payroll data is a way out for the Fed, and the Department of Labor's CPI data should follow the same pattern. Target time to prevent the Fed from raising interest rates in the near future.Tomorrow at 8:30 AM ET, the Bureau of Labor Statistics drops the July inflation read. Most traders will stare at the headline figure. The ones who profit will be watching what happens after the first five minutes of chaos. Here is what the tape is actually telling us right now: The Macro Backdrop June headline inflation came in at 3.5% year-over-year, down from 4.2% in May, with the month-over-month figure falling 0.4% seasonally adjusted. Core inflation held flat for the month. The 10-year Tre#现货ETF资金分化, BTC selling pressure remains
I frankly state: ETF funds flowing in and out is the market norm; there's no need to panic and be bearish just because of a few days of outflows; In the short term, BTC selling pressure is indeed accumulating, but its long-term foundation remains unshaken.
Recently, spot BTC ETFs have shifted from net inflows to net outflows, while Ethereum ETFs are still attracting small amounts of funds. Coupled with rumors of whales and miners continuously selling off, many friends have panicked and predicted the market will turn downward. In my opinion, this is an overreaction.
There are two reasons for this:
First, whether institutional ETFs or corporate treasuries, they never just buy without selling. Previously, when there were consecutive net inflows, no one loudly declared a bull market was established, but after just a few days of outflows, it was predicted to peak, which seems to be led by short-term sentiment. Having capital flowing in and out is the healthy market norm.
Second, the divergence between BTC and ETH is not due to capital withdrawing from the crypto market, but rather a rotation of on-exchange funds among mainstream assets. In the stock game, there is often a seesaw effect where one side gains and falls, so it's not about one side completely weakening.
My own operations remain conservative: the long-term bottom position remains firmly unmoved, neither rushing in after a few days of net inflow nor panicking in short-term outflows. From a short-term market perspective, trapped positions above and heavy selling pressure are indeed heavy; breaking through previous highs overnight is by no means easy; but the downside support is strong, so the room for deep declines is also limited. If you want to increase your position, wait for a pullback stabilization and clear signals before acting; that is the safest approach.
Cultivating the Dao should avoid impatience and impatience; trading should avoid chasing gains and selling losses. Let short-term capital flow in and outThe headline $1.1B weekly inflow masks a more useful signal: marginal demand is beginning to separate across the two assets. On Aug 10, Bitcoin ETFs saw roughly $91M of net outflows while Ether ETFs added about $5.3M.
For BTC, that matters alongside reported whale sales and miner transfers to Binance. ETF demand does not need to disappear for the balance to weaken; it only needs to absorb less of the available supply. CPI may support risk appetite, but sustained flow divergence would argue for watching market depth, not just cycle narratives. Not advice, just analysis.
#BTCETHETFFlowsDiverge#财报观察员: AI infrastructure earnings report debuts one after another
SpaceX's first batch of restrictions didn't fall below the price; instead, it rebounded to the IPO price. I actually think we need to be extra cautious here. Before the lock-up, the market was unanimously bearish, and the first wave was likely to be the first to sell short positions; On August 20, about 7% of the restricted shares were unlocked. After the short squeeze ends, whether they can catch the second round of shares is the real stress test.
What is even more worth watching this week is the AI industry chain financial report. Lumentum and Coherent are looking at 800G/1.6T optical communication demands; CoreWeave is looking at GPU leasing and AI computing power needs; Applied Materials is looking at fab capital expenditures; Cisco can verify whether enterprise AI network upgrades are truly starting to be spent.
Putting these financial reports together, they perfectly answer one question: have the massive AI Capex investments invested by Microsoft, Google, and Meta really translated into real orders for semiconductor equipment → computing power→ optical communications→ networks?
If these sectors continue to grow simultaneously, AI's current high valuation will at least have performance support; If orders slow down and profit margins decline, the market will start repricing the AI Capex cycle.One piece of data really resonates with me: the average cost base for BTC ETF holders is around $83,000.
And now BTC is at 65,000.
In other words, most institutional investors who bought BTC through ETFs are currently losing money. The average unrealized loss is 22%.
Do you know what that means? They have strong motivation to sell when the price rises back to around 80,000—because they've finally broken even.
This is known as "unwinding selling pressure."
Between 83,000 and 65,000, huge potential selling pressure has accumulated. Every time BTC rises a bit, a group of people say, "Finally broke even, sold it."
That's why after BTC rebounded from 58,000 to 65,000, it couldn't rise. It's not that there are no buyers, but that buying pressure was being eaten up by uneven selling.
My judgment: For BTC to truly start a new rally, it needs to effectively break through 70,000. Because above 70,000 is the real profit range for most ETF holders—above that, selling pressure actually decreases.
But 70,000? No catalyst in the short term is visible. Last week, Strategy sold 1,690 BTC at a "loss" at an average price of $64,262, using all of it to buy back STRC preferred shares, while also raising $650 million by issuing 6.58 million shares of MSTR common stock (six times the funds raised from selling coins); This seemingly abnormal loss-making operation is actually the company's capital operation shifting from "pure coin hoarding" to a "digital credit framework," aiming to maintain overall financial stability by increasing up to $4.65 billion in US dollar reserves to ensure payment of preferred stock dividends and debt interest.Key point: Some thoughts on this round of knockoff season:
If you play this cycle by the logic of the previous knockoff cycle, your mindset will inevitably collapse.
What people think of as the knockoff season:
1. Just like last round, all coins have risen, regardless of whether they're new or old
2. Only when the rate of increase matches the previous cycle is it truly a rally, and that's what makes it a knockoff season
Actually, many knockoff sectors in this bull market are already quite large. If you don't make money, it's because of that
1. Missing the right rhythm and missing out on surging sectors
2. Late entry and high chip costs
For example, if you buy a WLD at 9U or Ordi at 70U, you call it trash. Without a knockoff season, just look at how much they've risen from the bottom. Why didn't you buy earlier?
The two versions above are both part of the local hotspot season,
Investors holding onto old coins shouldn't expect all old coins to reach or even break previous highs—not every coin is inj
Most old coins have already exited the stage of history, but in the final stage of the bull market, they all make a shot to show respect for the bull market.#英伟达推动5000亿美元AI基建融资
Is AI entering the era of capital-driven infrastructure?
The next phase of AI may be about more than just chips
Nvidia, together with BlackRock, Blackstone, Goldman Sachs, and other institutions, is promoting an AI computing power financing platform worth over $500 billion, helping customers build data centers and purchase GPUs, accelerating the expansion $NVDA of AI infrastructure
AI is beginning to shift from investment by tech companies to infrastructure projects involving the entire capital market
The logic of the past
Cloud vendors buy GPUs, NVIDIA sells chips
But in the future, it may become
Financial capital provides funding → builds data centers → buys GPUs → leases computing power→ continuously generates cash flow
NVIDIA is gradually shifting from being a shovel seller to someone involved in the entire mining farm's construction. However, the fact that the market hasn't surged wildly also shows that investors are starting to focus on another issue: with such massive AI investment, where will the final returns lie?
After all, if trillions of dollars are invested continuously in the coming years, relying solely on companies like Microsoft, Google, and Amazon to pay out of pocket is unrealistic. So Wall Street's entry is actually a way to solve the funding bottleneck for AI development
But the risks are obvious: capital will accelerate AI development, which may also lead to oversupply. If computing power growth outpaces AI application monetization, the market may once again question investment returns. So going forward, we can't just look at how many GPUs NVIDIA sells
Key points to look at:
▶️ Are data centers profitable?
▶️Is the demand for AI cloud sustained?
▶️ Can the server industry, optical communications, storage, and power sectors benefit together?
I think this could signal that AI is moving from the first stage to the second stage
▶️ The first stage is to earn chip dividends
▶️ In the second phase, opportunities may spread throughout the entire AI infrastructure
The true future winners may not be just companies selling GPUs, but the entire AI infrastructure ecosystem
AI is transforming from a tech story into a super capital cycle
Non-investment advice for DYOR Wang Chun, co-founder of F2Pool, has recently spoken out, pouring cold water on the recent heated discussion in the Bitcoin community about the BIP-54 protocol upgrade. Statement: No support, but no dead end Wang Chun made it clear on X that he currently does not support the BIP-54 proposal. But he left a key "lifeline": if the proposal strictly follows the BIP-9 activation mechanism and reaches the required absolute majority signal threshold, the F2Pool team will update the mining pool nodes. Until then, F2Pool will not participate in the BIP-54 signal vote. What does that mean? To translate, Wang Chun's position can be broken down into three layers: Personal attitude: Opposes BIP-54, believes it's unnecessary. Bottom line principle: Respect the rules. If you complete the BIP-9 process and gain enough support, be willing to follow consensus. Signal: Will not actively 'take sides' before the standard is officially met, and won't create fake miner support for the proposal. This is actually saying: Don't try to take shortcuts, just follow the rules. Why oppose it? There were already foreshadowings. Wang Chun had already publicly expressed her reasons for opposition as early as April this year. In his view, several problems BIP-54 tries to solve—time warp attacks, block verification efficiency, transaction collisions—are not urgent. He even bluntly stated: "Bitcoin protocol upgrades should not be forcibly pushed forward like the 'bundle bill' by U.S. politicians." This statement hits the heart of the matter: he does not accept the logic of "packaging" escalations just to pass certain clauses. Miners' "obsession with rules" Wang Chun's statement reflects a core aspect of Bitcoin governance📌 BlackRock accumulated $470 million in a single week! $BTC Converging at the end of the triangle, a market reversal is imminent
According to Woofun AI monitoring, institutional funds are accelerating the reshaping of the Bitcoin market landscape, with massive accumulation in IBIT (IBIT.US) under BlackRock (BLK.US) becoming the core driver of this rally. Meanwhile, BTC has reached the end of a converging triangle on the 4-hour level, making the direction selection urgent.
On-chain data shows that this week, BlackRock transferred approximately $478.5 million through IBIT products, equivalent to 7,320 BTC, all transferred to the ETF-specific custody wallet. The largest single transaction involved 1,840 BTC (about $118.9 million), which was withdrawn from Coinbase Prime and transferred to cold wallets. Although the exact purpose of each transfer is still difficult to confirm, the weekly net inflow clearly demonstrates institutions' firm intention to build positions at the current price level, providing solid liquidity support for the market.
Technically, BTC is currently priced at around $64,964, right at the intersection of two trendlines in a triangle. The upper resistance line is pushing down from the July high of around $68,000, while the lower support line is rising from around $61,000. Both have narrowed, forming a typical momentum structure. Repeated price attempts have failed to effectively break through resistance, so the pattern remains neutral for now.
If bulls can break through this resistance line with increased volume, the $66,000 level is likely to turn from resistance into support; Conversely, if the lower boundary is breached, the price may pull back to the $62,000–$63,000 range, testing support below.
Regarding momentum indicators, the MACD histogram is positive but the difference is narrow, indicating insufficient upward momentum; The RSI recorded 61.86, slightly above neutral but not overbought, leaving some room for further upside. Overall, the direction selection is near, and the effectiveness of a breakout requires trading volume to confirm the situation.英伟达 @nvidia 宣布 与阿波罗、贝莱德、黑石、高盛、KKR等达成合作，筹集 $5,000亿 建立人工智能计算基础设施融资平台。
老黄也表示：最多兜底 25%，即 $1,250 亿。
翻译一下：
就是借钱给客户买自己NVIDIA 的卡，AI 这轮已经从「拿利润买卡」进化到「借钱买卡」了。
这套玩法是不是叫供应商融资，2000年演过一遍？🫡 $NVDA 油V型+黄金暴拉+白银天量+金银比68+DXY趴地——五个信号指向同一个结论：供给修复→通胀降温→降息加速。这是中期利好BTC的宏观底色，就等BTC自己从$64K供应区走出来了。ETH 在 OKX 独占跌幅榜（-2.1%），大饼才 -1.45%。
美股代币化更诚实：XSOXL（3x做多半导）-5.46%，XSNDK（3x做空纳指）+3.47%，科技在悄悄避险。
费率 +0.0014% 中性，没人为 ETH 弱势抄底；OI 10.67万 BTC 不动，大饼是真空不是托。
ETH 弱于 BTC 就两种剧本：altseason 前末跌，或大饼吸血。恐惧29+缩量79%，我偏前者，放量前不下注。
能拿走的一条——盯 ETH/BTC 汇率：不破前低横住=alt 蓄力；破前低下行=大饼独吞流动性。比盯横盘有用十倍。
你敢在 ETH 接刀吗？敢的评论说理由，不敢的说等什么信号（浅评被藏，带数字才前置）。
—— 链上老中医·每小时把脉 · 2026081113 · ETH失血
加密资产高风险，本文不构成投资建议，纯属个人观点。
#OKX星球 $ETH $BTC #ETH领跌 #避险情绪 #山寨分化存储股财报后遭血洗，AI内存牛市见顶了吗？
三星利润暴增18倍、海力士毛利率直逼70%、闪迪营收大涨3.7倍，但股价却集体下挫。这是一场典型的“完美预期下的估值修正”。
核心矛盾究竟是什么呢？
不是业绩不行，而是指引没能满足市场已被吊到天花板的胃口。闪迪下季度营收指引中值略低于最乐观的预期，年内暴涨数倍的高位获利盘借机集中出逃。
为什么说基本面还没崩？
“内存墙”刚需：大模型长文本推理对HBM、DDR5和企业级SSD的消耗是刚性的。
长协锁死产能：三巨头2026-2027年用于AI的先进产能已被云大厂通过长期协议（LTA）提前锁死。
寡头克制供给：三大原厂优先将传统产能切向HBM，传统消费级供给受限，供需紧平衡至少延至2027年。
必须警惕的结构性隐患：
严重偏科：完全靠AI数据中心续命。闪迪PC和手机存储业务同比大跌32%，传统电子消费市场至今未见实质复苏。
扩产周期的达摩克利斯之剑：目前各厂激进扩产的产能将在2027年底至2028年集中投放，届时若云厂商资本开支放缓，行业将重演过剩惨剧。
产业牛市逻辑未变，但单边暴涨的“讲故事阶段”结束了。股价剧烈回调是在挤压估值泡沫，等利空预期出清、估值回落到合理区间，砸出来的依然是更具性价比的二阶段右侧机会。
#存储股财报后下挫，AI内存牛市还稳吗？ #黄金挑战4380美元 The international precious metals market experienced an explosive rebound, with spot gold (XAU) strongly breaking through the $4,400 per ounce mark during trading, setting a temporary high.
This round of sharp gains is mainly attributed to the resonance of macro fundamentals and geopolitical expectations.
First, the recent U.S. July nonfarm payroll data unexpectedly fell short of expectations, significantly cooling market expectations for Fed rate hikes. The US dollar index weakened under pressure, directly benefiting precious metals denominated in dollars.
Second, rising market expectations for a easing situation in the Middle East have eased inflation concerns caused by rising energy prices, further weakening the need for central banks to tighten monetary policy.
In addition, the continued gold buying by central banks of major global economies (such as China's central bank increasing holdings for several consecutive months) has provided solid bottom support for gold prices.
From a market outlook, although gold prices have experienced a rapid rally in the short term and accumulated some profit-taking, there is a technical need 📉 for a correction. However, from a longer-term perspective, against the backdrop of ongoing global economic uncertainty and the start of a rate-cutting cycle, precious metals still hold strong allocation value. Therefore, the overall strategy recommends maintaining caution in the short term and maintaining a bullish 📈 outlook in the medium to long term.
For the more volatile spot silver (XAG) and the highly leveraged COMEX gold futures (GC), aggressive investors can wait for pullbacks and stabilization before buying on dips, closely monitoring subsequent US economic data and marginal changes in geopolitical dynamics.
$XAUThị trường trái phiếu của Nhật Bản đang phát đi một tín hiệu vĩ mô quan trọng khác."
"Lãi suất trái phiếu chính phủ Nhật Bản tiếp tục tăng, với lãi suất kỳ hạn ngắn đạt mức chưa từng thấy trong hàng thập kỷ khi thị trường định giá khả năng siết chặt chính sách từ Ngân hàng Nhật Bản. Lãi suất tăng thường phản ánh kỳ vọng chính sách tiền tệ siết chặt hơn, lạm phát kéo dài, hoặc lo ngại gia tăng về tình hình tài chính."
"Vấn đề quan trọng không chỉ là Nhật Bản mà còn là tác động toàn cầu. Lãi suất Nhật Bản cao hơn có thể khuyến khích các nhà đầu tư trong nước đưa vốn về, giảm nhu
cầu đối với tài sản nước ngoài và thắt chặt thanh khoản toàn cầu.
Điều này có thể ảnh hưởng đến cổ phiếu, trái phiếu và các tài sản rủi ro, bao gồm cả tiền điện tử."
Dù lãi suất tăng không tự động dẫn đến các đợt bán tháo thị trường, nhưng đây là một chỉ báo vĩ mô quan trọng. Nếu thị trường trái phiếu Nhật Bản tiếp tục đi lên, các nhà đầu tư sẽ theo dõi sát sao xem liệu nó có bắt đầu ảnh hưởng đến dòng tiền toàn cầu và tâm lý rủi ro rộng hơn trong những tháng tới hay không.
$XRP
#AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra The $6.4 billion crypto treasury plan terminated: what truly changed was not a single project, but the financing logic
Many people, upon seeing crypto cooperation terminate, immediately react with the "project failure." But what is even more worth studying this time is why a transaction that originally intended to tie a SPAC, a listed company, and a single token treasury was jointly pressed by all three parties after market conditions changed.
On August 7, Trump Media, Crypto.com, and Yorkville Acquisition signed a mutual termination agreement to cancel the business combination planned to establish Trump Media Group CRO Strategy. The SEC filing and the joint announcement of the three parties stated that termination was mutually agreed upon by all parties, citing changes in market conditions and business and stakeholder priorities. Related ETF service cooperation was also discontinued.
The original plan was not a routine partnership, but a digital asset treasury company centered on CRO: entering the public market through a SPAC, combining token assets, equity financing, and credit lines. This model connected token price, listed company valuation, financing capacity, and market sentiment within the same structure. During the upward phase, this combination amplified capital efficiency; When the environment weakened, various links may also increase pressure on each other.
Terminating the agreement does not prove that the digital asset treasury model has failed, nor does it mean that relevant participants are withdrawing from crypto business. It indicates that the market is beginning to redistinguish between "understandable narratives" and "transactions that can be completed." For any large treasury plan, investors need to observe whether funds are secured, how assets are valued, how lock-up periods are designed, whether financing relies on sustained premiums, and whether exit mechanisms are clear.
This may be an early signal: the competition in crypto treasury is shifting from announcing scale to balancing sheet quality and execution certainty. If market volatility continues, more projects will downsize, delay settlements, or rewrite terms in the future; If the financing environment improves, the model may return, but the structure will emphasize cash buffers, phased investments, and risk isolation.
Do you think the most critical factors in digital asset treasury are holding size, financing costs, or exit and risk isolation mechanisms?
#加密财库 #SPAC #行业观察 #Web3
Risk Warning: This article is for industry research purposes only and does not constitute investment, legal, or trading advice.Based on the 1-hour xSPCX/USDT (SpaceX Tokenized Stock) chart provided, here is a simple prediction and market breakdown:
Price Prediction
Short-Term Range: $XSPCX 135.00 to $139.50
Trend Outlook: Neutral to Bullish / Recovery
Key Market Details
Current Price: $137.03 (showing a small gain of +0.29% over the last 24 hours).
Support Level: Around $XSPCX 135.71 (marked by the MA20 moving average line) and $130.20 (the 24-hour low).
Resistance Level: Around $139.52 (the 24-hour high seen on the chart).
What This Means
The chart shows that after experiencing a sharp drop and a quick rebound, the price has stabilized and is moving slightly upward near the $137 mark. The short-term moving averages are flattening out and turning upward, showing a mild recovery in momentum. Expect the price to trade comfortably within this range as it tests higher resistance levels.#AIInfraEarningsWatch #OKX.ai 📉 STRATEGY DUMPS 1,690 BTC, ADDS $650M TO CASH RESERVES
Strategy (MicroStrategy) had a notable trading week, offloading 1,690 BTC to net about $108.6 million while selling over 6.5 million MSTR shares for $653.1 million. All of this capital was channeled into repurchasing STRC preferred stock and boosting its cash reserves to approximately $4.65 billion as of August 9.
This move shows that the market's largest Bitcoin holder is shifting from a pure accumulation strategy to balance sheet management. Although it still holds a massive treasury of 840,447 BTC, flexibly taking partial profits helps build a liquidity buffer to defend against sharp market volatility. Consequently, the firm is actively managing its capital flow to ease financial pressure rather than passively holding through downturns.
#StrategySellsBTCAgain
#BTCETHETFFlowsDiverge #英伟达推动5000亿美元AI基建融资
On one hand, $NVDA is partnering with institutional giants like BlackRock and BlackRock to leverage $500 billion in plans to build data centers
On the other hand, $15 billion in $INTC financing plans will support AI and advanced manufacturing
Technological competition has also extended to capital competition. If more funding and capital favor are obtained, from a certain perspective, there will be more opportunities to capture more markets and influence the overall future industry landscape
The investments of these two tech giants in AI also serve as a reassurance to ordinary investors' concerns about AI valuations or whether AI narratives can be realized and translated into actual performance demands
When giants raise funds, the AI market is vast and there is still demand to be developed. What requires caution is the existing market competition, aiming only to capture market share
If you continue to invest in the AI sector, the best choice is still to buy in batches when prices are lower, because these companies have both technical and financial advantages and a large market share, making them more certain
For AI companies with high growth potential, they should be classified as risky investments, which are high-risk, high-reward venture capitalists with relatively small share of capital and less capital advantage compared to tech giants
In terms of technology and capital, long-term market dominance ultimately leads to steady performance growth, efficient capital conversion rates, a solid technological moat, and real project business needs, which may lead in 🤔 long-term competition🚨 Strategy firm still has $785 million left to complete the final $5 gap for STRC
Strategy is continuing its aggressive moves, pushing its STRC preferred shares back up to the $100 target; The latest data shows how much capital is still being injected.
Under its $1 billion STRC share buyback program, the company still has about $785.2 million in funds remaining.
In the past three weeks, Strategy spent about $214.8 million to buy back 2.3 million shares of STRC. To fund these buybacks, the company sold about 3,328 BTC, amounting to approximately $213.3 million.
This means Strategy is leveraging part of its massive Bitcoin reserves to support its preferred equity structure and enhance STRC's market position.
📊 Key figures:
• STRC buyback program: $1 billion
• Remaining amount: approximately $785.2 million
• Repurchased STRC shares: approximately 2.3 million shares
• BTC sold: about 3,328 BTC
• BTC sale revenue: approximately $213.3 million
• Strategy's BTC holdings: 840,447
• STRC target: $100
The bigger story isn't just about Strategy selling Bitcoin.
It reflects the evolving structure of corporate capital. Strategy's reputation is built on continuous Bitcoin accumulation, but its expanding preferred equity ecosystem also brings ongoing financial obligations. Therefore, supporting STRC may become increasingly strategically important, as a healthier preferred stock market helps Strategy maintain access to capital.
For $BTC, this sale is relatively small relative to its size—because Strategy's holdings are extremely large. But investors will closely watch whether further BTC sales are needed to complete the remaining $785 million buyback capacity.
The most critical issue right now is:
While continuing to protect its long-term Bitcoin accumulation strategy, can Strategy fill the remaining STRC gap?
This is becoming an increasingly important part of the Strategy + Bitcoin reserve model.
#bitcoin #BTC #strategy #MSTR #strcToday's Bitcoin and Ethereum market analysis
A reminder to short sellers, don't panic. Don't think you can't go down. The reason you are grinding here is because there are no chips below, and the market makers are waiting here for chips.
Taking a trade or rebounding is just going short.
Short around BTC 64,300. Short near ETH 1,888
Prioritize shorting Bitcoin. $BTC $ETH #现货ETF资金分化, BTC selling pressure remains #英伟达推动5000亿美元AI基建融资
英伟达联合多家头部机构搭建AI算力融资平台，目标长期撬动超5000亿美元第三方资本，理论上可以加速算力集群建设、拉动GPU需求。
但消息公布后股价下跌约2.8%，市场核心顾虑：需求是否需要靠融资杠杆才能兑现，产业链循环融资风险抬升。
5000亿只是框架动员目标，不等于营收和现成订单。往后重点看两点：资金是不是真正新增、数据中心项目现金流能不能覆盖债务，这才是决定估值走向的关键。
$BTC $ETH $BICO 🇺🇸 US MACRO: The Clarity Act clock just started ticking
Here's the setup most people are missing:
On 8/8, the Senate teed up the Clarity Act for its most important moment yet — filing the procedural motion that clears the runway for a floor vote. It's not law yet, and it's not even voted on yet. That vote lands when the Senate returns from recess, Sept 14-16. That's the date that actually matters. It still needs 60 votes — all Republicans plus 7+ Democrats — and a handful of holdouts are still negotiating conflict-of-interest language before they'll sign on.
Zoom out and the picture is still bullish for the space: the House already passed its version last July, 294-134, with strong bipartisan support. The industry's central policy bet has survived every delay so far. September is the real test.
If it clears — reduced legal risk, institutional capital finds an on-ramp. The names positioned to benefit most from a compliance-first framework: $XRP, $ADA. $SOL's case is different — less "built for regulators," more "dominant venue for RWA and DeFi volume." Two different bets, not a ranking.
Meanwhile, the macro backdrop hasn't loosened: Fed funds sitting at 3.50–3.75% for five straight meetings under Chair Warsh, tone still inflation-first. Liquidity hasn't turned yet — that's still the short-term lever, not the long-term one.
And on Trump Media/Crypto.com — that was a mutual wind-down of one specific treasury vehicle, both sides citing a crowded market rather than a falling-out. Trump Media kept its existing CRO position and its separate 9,500+ BTC holdings; CRO dipped 5-8% on the news but this wasn't a clean exit.
My take: Clarity is the long game, the Fed is the near-term trigger. If Warsh cuts at the next meeting, BTC's current range (~$64K) becomes a lot more interesting to watch break.
What's your read — does Clarity clear the Senate in September, and who's better positioned: $XRP or $ADA ?
#AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra
$BTC Currently, the AI industry is experiencing a highly fragmented phenomenon of "ice and fire." On one hand, U.S. cloud companies are almost in an "arms race" with full firepower, significantly increasing capital expenditures every fiscal quarter and frantically purchasing computing infrastructure. This massive investment has even pushed many giants' net cash flow to the brink of negative territory, but every time a new quarter's financial report is released, they always present data showing "huge demand and supply exceeds supply" to persuade the market to increase investment. Driven by this, the global semiconductor hardware industry has soared, and even Chinese companies have indirectly taken a share. However, this dividend mainly revolves around the needs of U.S. cloud manufacturers and is essentially an extension of the U.S. AI supply chain. In contrast, China, despite occasional exciting news about domestic open-source large models and promising development, has shown particular restraint in infrastructure investment. So far, almost no domestic cloud provider has offered capital expenditures even close to those of their American counterparts, and the market has not experienced the "severe supply shortage" that troubles American peers. This raises an extremely unusual and thought-provoking core question: why does a superpower with 1.4 billion people and a comprehensive industrial system appear to have much less AI computing power demand than a financial powerhouse with only 300 million people? In fact, the fundamental reason behind this is not population or national strength, but rather the "misalignment" between the AI demand and business logic of China and the U.S. U.S. investment is a typical example of "selling shovels" logic, focusing on training billion- or trillion-parameter levels酷儿的叙事其实一直没变过——让比特币生息。
2024到2025年做了原生BTC质押，TVL干到过1.5亿美金。2025年12月发了《CORE营收路线图》，把2026年定位成“营收时代”，核心就一句话：砍掉通胀补贴，用生态手续费回购酷儿。
三大引擎：lstBTC流动质押、SatPay比特币银行、AMP资管策略。SatPay是今年最核心的产品，3月份开始Beta测试，用户质押BTC借稳定币，然后用借记卡消费。这条路如果跑通了，酷儿的估值逻辑会从“叙事驱动”变成“现金流定价”。但目前SatPay还没全民开放，回购也还没见到真金白银的链上记录。地基打好了，但商业闭环还没走完。$CORE #Strategy再卖1690枚BTC, corporate financial pools are diverging
SEC filings revealed that Strategy sold another 1,690 BTC, cashing out to repurchase preferred shares and increase its US dollar cash reserves, while its holdings remained above 840,000 BTC.
Although selling accounts for a small proportion of total holdings, the once benchmark company of "buy only, not sell" has officially switched to a dynamic treasury model where buyers can buy and sell, completely diverging the entire corporate Bitcoin treasury track.
✅ Multi-head perspective
This sell-off is part of corporate cash flow management, not bearish on Bitcoin. The overall holdings remain large, with no signs of large-scale liquidation.
Some corporate treasuries are still buying BTC on dips, indicating that the logic of treating Bitcoin as a corporate reserve asset has not been overturned, and long-term institutional demand still exists.
⚠️ Realistic risks that cannot be ignored
The narrative has changed: the market previously assumed the strategy was a perpetual buyer, but now it becomes a potential seller. Once dividend or debt repayment pressure arises, sustained selling supply will appear during the rebound phase.
Within the industry's treasury, there is clear divergence: some companies continue to hoard coins, others reduce holdings at high prices, and are no longer a unified bull camp.
Don't just look at sell volume; focus on signal impact. During a rebound, such news can amplify market panic.
📌 My personal opinion
This selling of coins does not mean the end of the bull market, but it does require removing the "faith filter."
The financial treasury of listed companies must first be accountable to shareholders and debts; Bitcoin is just an asset, not a faith chip that will never be touched.
Going forward, don't overanalyze single small sells, but keep tracking: selling frequency, US dollar debt pressure, and the real changes in BTC per share.
Compared to single reductions, high-frequency and continuous selling is the real risk signal.
Dial mapping:
In the short term, actual selling pressure is limited but will dampen market sentiment. Combined with ETF capital divergence, Bitcoin's push to the 67,000 resistance zone will be even more difficult. The real major direction will still be macro variables like CPI and US Treasury yields.