
Orbit Post Sitemap
🚨 Bitcoin ETF inflows abruptly stopped—August 10 recorded a net outflow of $144.67 million!
On August 10, the U.S. spot Bitcoin ETF ended a five-day inflow streak, with a net outflow of $144.67 million that day. Meanwhile, Ethereum ETFs also cooled down, recording a net outflow of $14.59 million, breaking the previous four-day inflow trend.
---
1. Data Overview: Both inflow records have ended
Variety Leading Trend August 10 Net Flow Signal Meaning
$BTC ETF saw net inflows for five consecutive days (about 850 million yuan) -144.67 million yuan, short-term profit-taking / pre-CPI risk aversion and reduced positions
$ETH ETFs saw net inflows for four consecutive days (about 240 million RMB) -14.59 million HKD. Institutional enthusiasm for chasing ETH has temporarily cooled
---
2. How should we understand this leak?
1. Sensitive timing — a typical defense before CPI release
Funds choosing to withdraw before the release of major data is a classic safe-haven reduction rather than a trend reversal. Institutions are reluctant to endure overnight uncertainty before the CPI release, especially since the probability of a rate hike remains around 44%.
2. Normal profit-taking after continuous inflows
Previously, BTC ETFs attracted over $850 million in funds for five consecutive days, accumulating a large amount of short-term floating gains. Realizing some profits before CPI is a healthy technical adjustment, not a panic flight.
3. Outflow scale is relatively controllable
· BTC: 144 million vs. 850 million inflows over the previous five days→ with a retracement rate of only about 17%
· ETH: 14.59 million vs 240 million in inflows over the previous four weeks→ with a pullback ratio of only about 6%
There was no panic stamping by funds, and it was closer to orderly reduction.
---
3. What to watch next?
(1) CPI Data (Wednesday Night)
· Below expectations → Today's outflows are highly likely to return to chase the rally, with BTC possibly reaching 68,000
· Higher-than-expected → outflows may continue, first looking at $BTC 63,000** and **$ETH 1,850 support stiffness
(2) Whether today's ETF data has reversed
If ETF data turns positive again on August 11, it confirms that yesterday's outflows were defensive moves before CPI, not a trend reversal.
(3) Degree of Volume and Price Coordination
Even if funds flow out, if BTC can stabilize around 64,000 on reduced volume, it indicates spot buying is still in place, and there is no need for excessive panic.
---
4. Operational response
· Not turning bearish due to single-day outflows—the macro narrative remains unbroken, and the ETF's long-term trend remains upward
· Don't rush to buy the dip—wait for CPI to land, confirm the direction, then follow along
· If BTC falls back to 63,000 and ETF inflows resume →, it is considered an increasing position
---
Conclusion
Single-day outflow is noise; continuous outflow is the signal.
The data for the week after CPI is implemented is the real trend confirmation window to watch.
---
#BTC #ETH #ETF #CPI #加密市场分析$NVDA
Jensen Huang is a man who knows how to set traps! This $500 billion AI financing could exceed a trillion in the future.
He led five schemes:
1. Turn $NVDA chips into assets
NVIDIA used to sell GPUs, but now chips + system + software are packaged into long-term rental infrastructure, like water, electricity, and coal.
2. If you don't pay yourself, let the top Wall Street tycoons take the blame
Goldman Sachs, BlackRock, BlackRock, KKR, Apollo, and Brookfield were all brought in. All 500 billion yuan is third-party long-term funds, with Nvidia offering zero guarantees and zero endorsements. Risks are borne by others, but the ecosystem is controlled by itself.
3. Turning supply-demand imbalances into cash machines
The computing power gap is now ridiculously large: demand for large models has increased sevenfold in half a year, and chips, memory, electricity, and land are all in short supply. AI token profits are shockingly high—whoever gets computing power can make big money!
4. Locking the entire upstream and downstream chain together
From wafers, HBM memory, packaging, and optical modules to power, nuclear power, data centers, and construction workers, everything is tied together. Customers use NVIDIA architecture to continuously upgrade their systems; the deeper the ecosystem, the wider the moat.
5. Pave the way for the next wave of physical AI and agent AI
Right now, digital AI is just getting started; intelligent agents are already exploding. Later, physical AI like robotics and autonomous driving will become the real big scene. Jensen Huang first opened up financing channels. When demand peaks, while others are still looking for money, NVIDIA already has a cash flow pool and clients.
NVIDIA not only sells shovels but also has Wall Street help sell them. This move is brilliant!News Interpretation | The US July CPI will be released tomorrow night, directly rewriting the pricing for the Fed's September rate hike, putting the crypto market facing a critical test
📌 Key Event: At 20:30 Beijing time tomorrow night, the US July CPI data will be released, which is the key indicator determining whether the Fed will raise rates in September. Currently, market expectations are highly balanced.
1. Current Market Pricing Status
The CME FedWatch tool shows: the probability of keeping rates unchanged in September is 55.6%, and the probability of a 25bp hike is 44.4%, basically close to a 50-50 split.
Currently, expectations are extremely different; even a slight deviation from expectations can quickly reprice market rate hike pricing, causing sharp volatility in risk assets such as the US dollar, US Treasuries, and BTC/ETH.
2. Data benchmark expectations
- Overall CPI year-on-year: expected 3.4%, previous 3.5%
- Core CPI year-on-year: expected 2.5%
Short-term variables: July's Middle East geopolitical conflict pushed oil prices back above $80, raising the risk of a rebound in energy sectors and posing risks to inflation, with data exceeding expectations and potential disturbances.
3. Three Major Scenario Simulations (Impact on Crypto Assets)
1. ✅ CPI Below Expectations (Inflation Cooling)
With a sharp downward revision in rate hike expectations, a weaker US dollar, and lower US Treasury yields, risk assets are benefiting from positive conditions, making BTC and ETH likely to see a rebound.
2. ➖ CPI meets market expectations
There are no signals exceeding expectations in the market, pricing remains unchanged, and the market continues to fluctuate within a range. In the short term, it is difficult to break out of a one-sided trend, with the focus mainly on digesting volatility.
3. ⚠️ CPI Exceeds Expectations (Inflation Rebound)
The market has raised the probability of a rate hike in September, hawkish expectations are rising, a stronger US dollar and liquidity tightening expectations are suppressing risk assets, and cryptocurrencies are under pressure and retreating.
4. Key points and risk warnings at the trading level
1. This data is driven by strong events, and volatility before and after data is highly magnified, significantly increasing contract trading risk. Heavy positions in gambling data are not recommended.
2. Oil prices and geopolitical factors are hidden variables in this CPI and cannot rely solely on expectations for one-sided forecasts.
3. CME probability only reflects market trading expectations and does not represent the Fed's final decision. Subsequent verification of policy paths will be based on PCE and employment data. Hello, I'm Dr. Bi. While the crypto market has long been accustomed to trading 24×/7, the traditional stock market is taking a crucial step forward. Nasdaq is about to implement 5-day continuous trading 23 hours a day, with only one hour reserved per day for clearing and maintenance. This institutional change is far more than simply extending trading hours; it will reshape the U.S. stock ecosystem and spread outward, affecting the overall trading logic of A-shares, Hong Kong stocks, gold, and even the crypto sector. Many retail investors' first reaction was positive news: Asian investors don't have to stay up late and can trade US stocks during the day during Beijing time. But peeling back the surface conveniences lies the reality of liquidity restructuring, amplified institutional advantages, and uninterrupted global risk transmission. Ordinary investors must understand the opportunities and traps brought by this transformation. First, clarify the system itself: after the new rules are implemented, except for the one-hour clearing window in Eastern Time, all other trading hours will be open. The new night trading window corresponds exactly to the daytime opening hours of A-shares and Hong Kong stocks. In the past, pre- and after-hours trading in US stocks was a niche supplement with limited trading volume, but in the future, night trading will be officially incorporated into exchange rules. Regulators have also imposed restrictions: market orders are prohibited during night trading, and limit orders can only be used to avoid extreme slippage risks under thin liquidity. However, this only alleviates the risks and cannot eliminate all hidden risks. Many people mistakenly believe that 23-hour trading means abundant liquidity around the clock, which is the biggest misconception. Liquidity does not double in sync with trading hours; the vast majority of trading volume remains concentrated in the traditional main trading hours of the U.S. stock market. The new nighttime time zone has sharply shrunk market depth,Strategy: "Selling Coins to Survive": Crisis Resolved or Delaying Tactics? True Rescue or Delaying Tactics?
In the past half month, Strategy sold 1,690 BTC at an average price of $64,000, cashing out $100 million. Including previous sales, the two transactions raised a total of $213 million, all used to repurchase STRC preferred shares. At the same time, selling common shares raised $653 million to supplement reserves, with a total of 6,948 BTC sold this year.
Currently, US dollar reserves stand at $4.65 billion, which Saylor says is enough to cover dividends and interest over the past three years. STRC preferred shares have rebounded from $74 to $95, just $5 below par value.
But concerns remain: BTC is still selling below cost, with ongoing paper losses; If STRC doesn't rise back to $100, it may need to keep selling; And the entire model relies on the assumption of a 10% annual Bitcoin increase; deviations would lead to recurring risk.
Conclusion: The liquidity crisis is temporarily delayed for two to three years, but the real relief will require STRC to return to face value and restart BTC buying. The market's view of "selling at the bottom" may be coming true.
$BTC
#Strategy再卖1690枚BTC, corporate financial pools are diverging
#现货ETF资金分化, BTC selling pressure remains 链上有个大户今天出货了,地址bc1qdj。
这个地址在6小时内,把1274枚BTC拆给了三家OTC台子:Cumberland、FalconX、Galaxy Digital,加一起大概8150万美元。
我看到有个细节值得说一下。它走的是OTC,不是直接在交易所挂单砸盘。OTC单子不摆在盘面上,盘口上不容易看出有人在卖,这种卖法通常是卖家想清仓、又不想太显眼。
至于这个地址背后是谁、为什么要卖,链上目前只能看到这些信息,别的看不出来。更多的结论,得等后续地址流向或者新闻出来才能判断。
不构成投资建议。$BTC #Strategy再卖1690枚BTC, corporate financial divergence Saylor was hinting a few days ago that it might continue buying BTC. Latest result disclosure: Between August 3 and 9, Strategy sold another 1,690 BTC, with an average transaction price of about $64,262, cashing out $108.6 million. But I won't simply interpret it as "Saylor bearish on BTC." Because the money wasn't used to run away, but entirely for the repurchase of STRC preferred shares. After selling, Strategy still holds about 840,400 BTC. citeturn438268search1 This indicates a very important change: corporate BTC treasuries have shifted from "buying only, not selling" to active management. Even more interestingly, during the same period, Trump media disclosed that he already held about 14,139 BTC, worth nearly $900 million. citeturn616665search1 So now is not the time for "companies to start fleeing BTC." Instead: some buy, some sell, some use BTC to adjust their capital structure. BTC is currently around 63.9K. My execution was simple: hold at 63.5K—64K, and I won't go short because of Strategy selling. It regained the 65K and broke through 66K, and I remained bullish. If even 63K can't hold on, then I'll be wary of corporate Treasury sell-offs + ETF fund differentiation starting to create real pressure#英伟达推动5000亿美元AI基建融资
#英伟达甩出5000亿AI融资大蛋糕, their own stock price reversed and dropped 2.86%, with $70 billion in market value evaporating in a single day, and the five-year CDS soaring 5.3 basis points. Wall Street was both joining the hype and throwing money around, secretly sweating nervously. To put it bluntly, Huang wasn't paying out of pocket; he was pulling in BlackRock, BlackRock, and the six asset management firms to build lending platforms, lending money to companies crazily buying graphics cards and building data centers. Rumors spread that they were issuing large mortgages for the AI industry, and the market feared a revolving leverage bubble. In our crypto world, decentralized computing power coins like RNDR and TAO have surged in short-term turnover to ride the hype, but massive amounts of traditional funds are flowing into centralized computing infrastructure, diverting new growth in the crypto sector in the short term, and the market remains volatile. I've always been cautious—I don't rush to chase short-term AI counterfeit moments, wait for industry positive factors to materialize and digest them, and then look for reliable opportunities, confidently that the bull market will gradually return.
#财报观察员: AI infrastructure earnings report debuts one after another
$ETH
This represents only personal views and is not investment advice.$ARB 价格在 3.20 美元附近横盘,30.00 亿美元流通市值与 200 万美元年收入形成鲜明拉锯。
73.1% 的流通比例配合 8000 万美元的 24 小时成交额,整体换手率低迷,买盘在当前位置表现谨慎。
高达 1500 倍的 P/S 倍数与 2026 年第四季度追加 3.50% 的解锁预期,持续对中远期估值形成压制。
缺失销毁机制导致的价值捕获不足,使现有的低换手状态难以直接消化 42.00 亿美元 FDV 带来的溢价。
向上破局需要周度手续费出现量级跃升以重构估值中枢,但若代码提交持续保持在 90 天 60 次的低频状态,上行逻辑将宣告失效。
若日成交额萎缩至 8000 万美元下方,估值出清可能驱动价格向较当前市值下寻 30% 至 50% 的区域探底,强价值捕获机制超预期上线则是该路径的终止信号。
当顶部趋势线被有效放量突破且成交额翻倍时,市场将从基本面定价彻底转向资金流动性博弈,估值承压的推演也会随之被证伪。
未来 7 天最值得关注的变量,是 $ARB 的链上周度协议手续费累计速度与 24 小时成交额能否突破 8000 万美元关口。
#英伟达推动5000亿美元AI基建融资 #标普收盘再创新高,8000点预期升温 #存储股抛压缓和,AI内存牛市还稳吗?When a very busy big shot suddenly posts a long article, there’s usually a reason.
First, Huang went on Twitter to post, then Zuckerberg wrote a 6,500-word long article in one go. Recently, both have been emphasizing the same thing: open source.
Yesterday, Zuckerberg published a long article titled "The Future Belongs to Everyone," and on the same day, Meta open-sourced the new model Muse Glimmer, also previewing that stronger model weights will be released soon.
The core of the long article is one sentence: superintelligence should not be concentrated in the hands of a few companies; it should be distributed to everyone, becoming a personal super assistant.
It sounds great. But looking at Meta’s overall actions, it’s very clear.
Open-sourcing model weights means actively lowering the barrier for others to make products;
At the same time, they continue to lock in computing power, glasses, social data, and advertising systems. The easier it is for others to use your model, the harder it becomes to bypass your ecosystem for monetization in the end.
This is completely opposite to OpenAI and Anthropic’s insistence on closed source and emphasis on "safe and controllable" approaches. Both sides use the "safety" narrative to serve their own business models: one says centralization is safe, the other says decentralization is safe.
The louder they say "The future belongs to everyone," the more you need to see clearly who is defining the rules and who is collecting the toll.#AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra 📉 Is memory prices stopping? Apple steps in, supply and demand shift!
Is memory prices stopping rising? The reason behind it has been found—Apple is testing Changxin Memory's DRAM chips.
Understand it in one sentence: Apple is looking for a "backup" for itself
Previously, Apple's memory relied entirely on Micron, Samsung, and SK Hynix. Now Apple has begun testing Changxin's DRAM chips, preparing to use them in iPhones and MacBooks.
What does this mean? Apple has bargaining chips in its hands—you raise the price? I'll replace you.
The three major manufacturers are hesitant to raise prices easily, which is one of the core reasons why memory prices have stalled.
There is another key point: China's memory chips are truly breaking through
Changxin's inclusion in Apple's testing list indicates that its product performance has reached a certain level. Once the tests pass and the U.S. government does not block it, Changxin may claim a share of the market.
The market's assessment of tight DRAM supply and demand has started to loosen, and related storage stocks (such as Micron, SanDisk, etc.) have shown weak performance these days, reflecting this logic taking hold.
Note: Although it hasn't officially launched yet, expectations have already shifted
It is still in the testing phase, and regulatory approval remains unknown. But the capital market is most sensitive—once news comes out, expectations change first.
Apple wants supply chain security + price-cutting chips, while Changxin wants endorsements from major clients. If passed smoothly, it would be good for Apple but negative for Micron and Samsung.
In short: memory prices can't rise—not the market's fault, but Apple having one more choice. The wolf has arrived—the anticipated story affects the market even before actual implementation 🍎
#苹果测试长鑫存储芯片并展开初步供货谈判 $ARB The current price of $3.20 corresponds to $3.00 billion in market value and $4.2 billion in FDV, while $2 million in annual revenue results in a P/S ratio as high as 1,500 times. The stark fundamentals put pressure on the price structure.
On the board, 73.1% of circulation is accompanied by $80 million in 24-hour turnover, with turnover rates remaining low, indicating a lack of capital interest near $3.20. The core variables driving the current valuation reversal are: a 2100x FDV-to-revenue ratio, lack of value capture due to no burn mechanism, and an additional 3.50% unlock expected in Q4 2026.
The condition for breaking upward is a scaling breakthrough in protocol revenue or a restructuring of the token mechanism. If on-chain fee inflows increase sharply or the buyback mechanism is implemented, the valuation center will be quickly flattened, and the price will break above the liquidity-dense zone. The trigger variable for this scenario is a significant weekly fee spike, while the failure signal is a low-frequency state of code submissions maintained 60 times over 90 consecutive days.
The downward support scenario is dominated by high-valuation clearance. If spot buying cannot absorb the premium from $4.2 billion in FDV, the price will test the support range downward, seeking a value rebalancing zone of 30% to 50% below the current $3.0 billion free float market. The trigger variable for this scenario is daily turnover continuously shrinking below $80 million, with the failure signal being the strong value capture mechanism launching beyond expectations.
The overall failure point of the price structure is set at the top trendline, forming an effective volume breakout. If $ARB is forcibly pushed up with doubling turnover and stabilizing at a high valuation, it indicates that the market logic has shifted from fundamental pricing to pure liquidity competition, and the original valuation pressure simulation will fail.
In the next 7 days, focus on $ARB's weekly protocol fee accumulation speed and whether its 24-hour transaction volume can break through the $80 million mark.
#三星钱包将接入稳定币, payment scenarios continue to expand. #苹果测试长鑫存储芯片并展开初步供货谈判 #本周三CPI公布, will the pricing for a rate hike in September be rewritten?"Say Things Without Saying Anything, Say Things Hard When Nothing Matters"
Gold· BTC·US Stocks | Analysis of hidden cross-market logic after nonfarm payroll implementation
🦅 The weakening nonfarm payroll data should not be crudely interpreted as a broad rise in all asset categories as positive news; the divergence across markets has long hidden different logics.
After the data was released, funds concentrated on expectations of Fed rate cuts, gold surged and held its highs, while US tech stocks remained volatile; BTC and ETH showed weak short-term gains and then pulled back again. Many people judged a weak dollar to go long across all assets, but the actual market trend was far more complex than this simple logic.
🐢 Gold is currently holding above the 4370 level. Global gold ETF capital inflows combined with long-term gold purchases by central banks provide a solid and stable medium- to long-term allocation logic. However, short-term profit-taking positions are heavily accumulated, with strong selling pressure on chips in the 4420–4450 range. Blindly chasing high prices now is not cost-effective and carries high risk. BTC and ETH may benefit from rate cut expectations, but they are high-volatility, high-beta risk assets, and their operating logic is inherently different from gold. Rate cut expectations essentially reflect a downturn in economic sentiment. If subsequent CPI inflation data remains high, the market will quickly lower rate cut expectations, putting pressure on risk assets like cryptocurrencies to fall first.
In summary: The expectation of rate cuts is a positive logic for gold, but it is a double-edged sword for US tech stocks and cryptocurrencies.
Only when the economy achieves a soft landing can the stock market and cryptocurrencies rally in tandem; Once the economy accelerates downward and recession trading logic dominates the market, risk assets will face concentrated sell-offs, with only gold's safe-haven value emerging independently.
🔴 At this stage, the core turning point of the market is the upcoming CPI inflation data, which will directly reshape expectations for the Federal Reserve's monetary policy pricing:
1. Inflation data has declined→ rate cut expectations confirmed, and US tech stocks, BTC, ETH, and gold have all seen a recovery rally;
2. Inflation data is stronger than expected→ dampening rate cut expectations, causing short-term pressure on gold to pull back, and the downward pressure on risk assets will multiply.
The market has already overused the market to gamble on the non-farm payroll data. Don't rely on the non-farm payroll results as a reference for subsequent trading; from now on, all focus must be on the CPI.
Currently, the overall market is in a volatile pattern. Avoid heavy positions and one-sided bets on price movements. Preparing two market scenarios in advance is far more reliable than subjective prediction of price movements. $ETH #本周三CPI公布, will the pricing for September rate hikes be rewritten? Review of the U.S. tech sector: Optical communications, storage, and semiconductors fluctuate in rotation, with sharp divergences in the AI main theme
$SNDK $SPCX
In the past half month, the AI technology sector in the US stock market has completely ended its one-sided rally. The three major segments—optical communications, storage, and semiconductors—have taken completely different roller coaster rides. Policy stimulus, earnings expectations, and profit-taking have all dominated the market, intensifying sector differentiation and heightening the atmosphere of capital competition. Today, we have thoroughly reviewed the current market flow and its underlying logic.
1. Optical Communications: Policies Trigger Short-Term Surges, Rapid Pullback After Good Signs Materialize
Starting in late July, demand for AI computing power bandwidth led the optical communications sector to steadily rise. From July 22 to 27, COHR and LITE closed higher consecutively, with the prosperity of demand for 800G optical modules providing the sector's underlying support.
The turning point in the market came in early August: news that the U.S. was preparing to restrict imports of Chinese transceivers early in advance, and domestic manufacturers saw an epic surge.
On August 4, the sector saw a peak rally: AAOI surged 21.7% in a single day, COHR surged 14.12%, LITE rose in tandem, and the Optical Communication ETF (FOTO) rose over 11% in a single day; The positive news continued into August 5, with the sector continuing to rise strongly.
The carnival lasted only five trading days. On August 10, the market fully digested policy dividends, short-term profit-taking sectors concentrated and pulled back, with LITE plunging 5.54% to lead the decline, while AAOI, POET, and Corning all closed lower, showing short-term leading signals from a technical perspective.
Core logic summary
Short-term market trends are fully tied to favorable geopolitical policies, but selling pressure has already emerged; Long-term fundamentals remain solid, with the expansion of AI data center computing power driving the popularization of 800G optical modules and the commercial rollout of 1.6T products. Demand for high-speed optical modules has long-term growth potential, and subsequent stock prices will need solid orders and performance to support the price.
2. Storage Sector: Extreme bullish and bearish tug-of-war, with surges and drops in and out of control
The storage sector experienced the highest volatility among the three major sub-segments, perfectly demonstrating the tug-of-war between "cyclical upward expectations vs. profit-taking panic":
1. First round of rally (late July)
Lam Research's earnings report far exceeded expectations, confirming strong demand for semiconductor equipment. Coupled with a significant increase in Microsoft Azure cloud capital expenditures, the market is confident that the storage chip cycle has bottomed out and is recovering. On July 30, the sector exploded across the board, with SanDisk soaring 23% in a single day, while Micron and SK Hynix both rose over 16%, with short-term gains that overshadowed much of the optimistic expectations.
2. Cliff-like Drop (August 6)
Optimism quickly reversed, and funds concentrated to take profits. The market began to worry that storage price increases were unsustainable, AI server demand was overdrawn early, and weak consumer electronics were dragging down end-user demand. Leading company Western Digital plunged 19% in a single day, SanDisk plunged 13%, and the sector instantly turned into panic.
3. Slight Oversold Recovery (August 7)
After the plunge, some funds bought the dip and entered the market, leading to a technical rebound in the sector. Micron, Western Digital, and SanDisk closed slightly higher, testing key support levels and stabilizing in the early stages.
On the fundamental side, DRAM and NAND chip prices will continue to recover from the second half of 2025, with AI computing power expansion as the core driving force; However, short-term valuations have fully priced into the upcycle, and future price movements will be closely tied to chip prices and cloud vendor purchase order data.
3. Semiconductor Sector: Sector Fragmentation Is Obvious, Performance Determines Strength
The semiconductor market as a whole has experienced repeated fluctuations, with significant differences in performance among different manufacturers:
In mid-July, the market was suppressed by concerns over extended Fed rate hikes, causing the Philadelphia Semiconductor Index to plunge into a technical bear market; On July 28, pessimism peaked, and the index plunged more than 5% again.
On July 30, Lam Research's better-than-expected earnings report completely reversed sector sentiment, with the Philadelphia Semiconductor Index surging over 6%, ARM and AMD surging sharply, and industry confidence briefly rebounding.
Entering August, the sector has become cautiously diverged:
• Intel and AMD benefited from improved PC and server demand, with significant increases on August 4;
• Nvidia's gains were weak, closing slightly higher, with funds beginning to favor lower-priced chip stocks;
On August 7, market risk appetite slightly rebounded, ARM rose slightly, and the sector mainly experienced weak oscillating recovery.
Overall, the semiconductor industry has emerged from the bottom and is recovering, but the pace of recovery is uneven: the device side is the strongest, followed closely by storage, general-purpose chips are recovering slowly, and the Fed's interest rate trend remains the biggest macro variable suppressing the sector.
4. Overall Market Outlook Summary
1. Short Term: All three major sectors have undergone a round of strong rally + pullback, ending the one-sided market trend, with volatile games becoming the main theme and significantly reducing the cost-effectiveness of chasing highs;
2. Mid-term: AI computing power remains the main theme throughout the year, with rigid demand for optical modules, memory chips, and computing power chips;
3. Risk Points: Repeated geopolitical policies, delayed Fed rate cuts, AI capital spending falling short of expectations, and pressure to realize profits at sector highs.
⚠️ Risk warning: This content is only a market review and summary and does not constitute any practical investment advice for buying or selling stocks. U.S. stocks are volatile and highly volatile, so be sure to manage your position risk properly. $BTC #现货ETF资金分化, BTC selling pressure remains
ETF inflows, strategies selling coins, rising oil prices—any single piece of news today could tell a completely different BTC story.
This is also the easiest part of market slacking: first have a conclusion, then pick a pleasing news story to prove yourself.
There's nothing wrong with news itself; the problem is that we often let it take on tasks beyond the scope of information. Discipline isn't just about restricting operations, but also about constraining interpretation: whatever a message can reveal, let it go as far as it goes.If you still understand $OKB today by "OKX platform coins, discounted fees, and periodic burning," I think we are already at least one version behind. That logic hasn't completely failed, but it's no longer the core variable determining $OKB's next stage of valuation. The truly noteworthy changes occur at two point in time. The first is August 2025. OKX burned 65,256,712 $OKB at once, permanently pushing the total supply to 21 million. Meanwhile, $OKB gradually migrated to X Layer and became the only native gas asset on X Layer. On the supply side, the dynamic model of "continuous buyback and continuous burning" has shifted to a model so simple it's almost brutal: total supply of 21 million, with no subsequent inflation. The second timeline is in 2026. X Layer launched Exchange OS. Its significance is not that "OKX has built another chain," but that it is beginning to attempt to bring the most valuable aspects of the exchange's capabilities—matchmaking, unified accounts, margin, risk management, market creation—into on-chain financial infrastructure. More importantly: to deploy and operate your own marketplace on Exchange OS, you need to stake $OKB. This changed the entire valuation logic. Previously, holding $OKB was mainly to share rights within the OKX platform ecosystem. What OKX wants to do now is make $OKB the economy needed to enter an on-chain trading infrastructure📊There is a paradox going on:
$BTC is down only 1.5% to $63,900, $ETH stands around $1,870
But the account for the majority of the brothers remained red, while a group of altcoins quietly exploded:
$CYS +42% for 24h,
$RAD +31%,
$CRV +11%,
$MNT +13%
Last week, the market was not short of money, it was extremely picky: gold $PAXG $XAUT increased by nearly 9%, $XMR +9%, and $SHIB -10%, $CRO -14%, $ONDO -9% — memes and RWA bleed, AI-infra, DeFi, privacy sucked up cash flow.
One perspective says that this is a risk-off signal, the altcoin is still bottoming.
But from another perspective, liquidity still rotates very quickly between stories — $SUI, $TAO, $PENGU, $DOS are all in the top trending today.
In my opinion, don't guess the bottom, look at where the cash flow is standing.
Maybe Altseason doesn't disappear, but just breaks into small waves — this week's winner isn't necessarily the year-end winner.
If you can only hold 1 altcoin until the end of the year, do you choose $MNT, $CRV or $XMR?
#CPIToResetFedBets #BTCETHETFFlowsDiverge #StrategySellsBTCAgain The rise of UB
UB (Unibase) Core Positioning
Unibase's native utility token, launched in September 2025, positioned as a decentralized AI memory layer, providing on-chain AI agents with long-term memory storage, cross-platform interoperability, and data sovereignty guarantees, aiming to build an "open agent internet" infrastructure, with the track belonging to the AI+Blockchain intersection field.
1. Key Basic Data (Latest as of August 2026)
- Token standard: BNB Smart Chain BEP-20, multi-chain deployment (Ethereum, Arbitrum, etc.)
- Total supply: 10 billion coins (fixed cap)
- Circulating supply: about 2.5 billion tokens, with a circulation rate of only 25% (75% unlocked)
- Current price: approximately $0.136-0.138 (24-hour increase of about 6-8%)
- Market capitalization: approximately $340–350 million, ranking approximately 118th among global cryptocurrencies
- 24-hour trading volume: about $200-300 million, moderate liquidity
- All-time high: about $0.199 (currently down about 31% from the all-time high)
2. Core Mechanisms and Value Support
1. veUB Governance Model (ve(3,3) variant)
- Locking UB to obtain veUB voting rights certificates with a maximum lock-up period of 4 years, with voting rights positively correlated with the lock-up period
- veUB can be used to vote on ecosystem reward allocation and protocol parameter adjustments, increasing staking yields
- Forming "UB Wars": Institutions compete for control of veUB to influence reward flows
2. Core functions and ecosystem applications
- Protocol fees: Pay for AI agent deployment, memory storage (Membase), and cross-proxy communication bandwidth fees
- Node staking: Node staking will launch in Q1 2026 to maintain network security and earn rewards
- Knowledge Mining: Contribute AI knowledge graphs to earn UB rewards
- Ecosystem projects: BitAgent (multi-agent collaboration), TradingFlow (autonomous trading agent), TwinX (self-evolving agency platform), etc
3. Token distribution and unlocking mechanism
- Community and ecosystem 45%, team and advisors 18%, treasury 15%, ecosystem 10%, early supporters 12%
- All tokens unlocked at 0% TGE, with a 6-month lock-up period followed by linear unlocking over 24 months, creating significant long-term unlocking pressure
3. Project Highlights (Relative Advantages)
1. Track Alignment with Hot Topics: AI agent infrastructure is currently a hot topic in the crypto market, addressing pain points in AI long-term memory and cross-platform collaboration
2. Unique Technology Positioning: Focused on the niche of AI memory layer, it forms differentiated competition with general AI platforms
3. High Exchange Recognition: Listed on major exchanges like Binance Alpha, Gate, and Bithumb, ensuring guaranteed liquidity
4. Rapid ecosystem expansion: AI frameworks such as MCP, Virtuals, and ElizaOS have been integrated, with ongoing ecosystem projects
4. Fatal Core Risks (Top Priority)
1. Extreme Dilution Risk (the Most Deadly)
- 75% of tokens remain unlocked (about 7.5 billion), with releasing over the next two years, creating huge selling pressure
- Fully diluted valuation (FDV) reaches $1.36–$1.38 billion, with current market capitalization at only 25% of FDV, making price support fragile
- Historically, there have been single-day drops of 30%, and the unlocking period may trigger even greater volatility
2. Governance and chip concentration risk
- Early investors and teams held large amounts of unlocked tokens, indicating significant control risks
- A few large holders can control governance through veUB, raising concerns about decentralization
- Some core team members have left, affecting the long-term stability of the project
3. Competition is intensifying
- Facing potential competition from traditional AI giants like OpenAI and Google, these giants can quickly replicate similar functions
- Directly compete with multi-track projects such as storage, data availability, and proxy platforms, resulting in dispersed resources
- Severe homogenization in the AI infrastructure sector, with Unibase lacking absolute technical barriers
4. Business models are not validated
- Core revenue depends on protocol fees; currently, the AI agent market is small in scale, and actual income is negligible
- The project is still in its early stages, with product implementation and user growth falling short of expectations, resulting in weak value support
- Highly dependent on AI narrative hype; once market heat fades, prices may pull back sharply
5. Security and compliance risks
- There are numerous phishing websites (such as Unlbase[.] app), which steal user assets
- There is regulatory uncertainty regarding token classification, and in some regions, it may be classified as a security
- Smart Contract Risks: The AI memory layer involves complex data interactions and poses potential vulnerabilities
5. Market Trends and Objective Conclusions Going Forward
1. Long-term fundamentals: highly speculative, extremely risky
- Advantages: AI sector popularity, exchange support, ecosystem expansion
- Disadvantages: Diluted bomb, centralized governance, intense competition, unproven business model
- Best Expectations: If the AI agent market surges, it may reach the $0.2-0.3 range; if selling pressure is unlocked, it may fall below $0.05
- Worst-case scenario: Project progress falls short of expectations, AI narrative cools, prices remain prolonged sluggish, or even gradually drop to zero
2. Short-term potential
- Positive catalysts: AI sector rebound, new exchange launches, major ecosystem collaborations, and increased veUB staking rate
- Market nature: Purely speculative pulse market, lacking sustained upward momentum, prone to rapid pullbacks after rebounds
- Key support: $0.10-0.12 range; Key resistance: $0.18-0.20 range
6. Summary and reminders
UB is an early project token in the AI+Blockchain intersection domain, with a concept that aligns with market hotspots but carries extremely high risks:
✅ Hotspots in the sector, support from exchanges, and potential for ecosystem expansionFundamental Research Report $ARB / Arbitrum (L2/Sidechain) $3.20
First, the conclusion: Arbitrum ($ARB) has an overall score of 60/100, with a narrative that emphasizes implementation over implementation. Looking at the three layers, the company team has cash reserves, the protocol network already shows signs of paid usage, and token capture has been implemented.
Arbitrum (token $ARB), L2/sidechain track. Focuses on ETH L2 leadership, Optimistic Rollup. Benchmarks OP and ETH. Traditional inter-enterprise collaboration relies on cloud servers and contract reconciliation; during high concurrency, gas surges, TPS is limited, and cross-chain bridge security incidents occur frequently. Public blockchains use unified state machines for trustless settlement, reducing reconciliation costs. Average order value is $50-500/month, requiring USDC or fiat currency settlement. Narrative-driven track, bear market usage cut by 60-80%. Positioned as an end-to-end vertical platform. Product launch: protocol layer officially operational, on-chain dashboard shows protocol fees accumulating, showing signs of paid usage. Latest version not found, 60 valid submissions in the past 90 days.
On the user side, address MAU is not disclosed, DAU is not disclosed, 24h transaction volume $80.00M, TVL not found. Wallet addresses do not equal monthly active individuals; concentrated holdings of large addresses overestimate actual user numbers. On the revenue side, user fees are undisclosed; supply-side revenue is about 80-90% of user fees (belonging to LPs and nodes), protocol treasury revenue is $2.00M, token holders buy and burn annualized without a burn mechanism. 24h transaction volume is business turnover, not revenue. Company profits do not equal protocol profits, protocol profits do not equal token holders profit. Code side: 60 valid submissions in 90 days, 25 active contributors, latest version not found. GitHub is A-level evidence that can be directly verified. Investment background: For company equity financing, look at PitchBook/Crunchbase (A-level); for token private and public funding, refer to whitepapers, release curves, and on-chain unlock contracts (A-level); market makers and ecosystem funding are B-level but do not represent long-term holdings by tech VCs; for technical integration, see API/SDK access evidence (B-level); strategic partnerships and logo walls are D-level. Using NVIDIA GPUs does not mean NVIDIA investment, and going public on exchanges does not mean strategic investment.
On the token side, total supply is 1,300,000,000, circulating 950,000,000 (73.1%), FDV $4.20B, next unlock 2026-Q4 (accounting for +3.50% circulating circulating), annualized burn buyback no clear buyback burn. Must you buy coins to use the product? Yes, strong value capture (Gas/Collateral/Service Access). Looking at it together with peers (unified caliber, no cross-sector random comparison): Arbitrum $3.00B, OP undisclosed, ETH undisclosed. FDV: Arbitrum $4.20B, OP undisclosed, ETH undisclosed. Annualized revenue: Arbitrum $2.00M, OP undisclosed, ETH undisclosed. Monthly active addresses or users: Arbitrum undisclosed, OP undisclosed, ETH undisclosed. Figures are based on public data snapshots; some omissions will be supplemented by official self-reported or industry standards. Valuation: circulating market cap $3.00B, FDV $4.20B, P/S 1500.0x, FDV divided by revenue 2100.0x. Pessimistic outlook: $3.00B at 50-70% of the original price, trading in a neutral range; optimistic outlook: revenue doubled, burn landed, enterprise clients entered, FDV would align P/S with the top players. To wrap up: solid fundamentals (score 60/100). Token value capture has been implemented (buyback/burn/gas). Circulating market cap is relatively high compared to fundamentals, overdrawing expectations, and FDV is moderate. Main risks: short-term large unlock dump, long-term protocol revenue wiping out, token demand relying solely on incentives (incentive interruption means usage collapse). Next, focus on these numbers: protocol fee weekly, burn amount, active address retention, TVL/loan balance, GitHub version release. The above is the logic and judgment of public information and does not constitute buy or sell advice. Core financial indicators deviate by more than 30%, conclusions need to be re-evaluated.
That's all for the fundamentals; leave the rest to the market.
#基本面研报 #加密 #研究 #OKXOrbitSupposedly,
Fort Knox holds 147m oz of gold.
Nobody has fully audited it since the 1950s.
How long would it actually take?
147M oz ÷ 400 oz per bar = 368k bars
Each bar needs 3 things
1. Weigh it
2. Ultrasound it (tungsten fakes weigh almost exactly the same)
3. Log it
3min per bar.
About 18.4k hours of total work.
Bitcoin audits its entire coin supply with one command under 60 seconds.
Don't trust. Verify.#AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra #本周三CPI公布, will the pricing for a rate hike in September be rewritten?
Three seemingly contradictory pieces of information appeared in the market today: $BTC and $ETH spot ETFs combined net inflows of about $1.1 billion over the past week; Strategy sold 1,690 BTC last week; Oil prices rose about 5% due to geopolitical risks.
If you pick only one piece of information, it's easy to end up with completely opposite conclusions. Bulls focus on ETF inflows, bears focus on companies selling coins, and macros focus on oil prices and CPI.
Part of trading discipline is not letting a single piece of news complete all the reasoning for you. Before the news affects whom, for how long, has the price been digested, and what forces are still hedging, news is only material before these questions are answered.
The more lively the market, the more you need to avoid being swayed by headlines about your opinions.🚨 $1.1B JUST FLOWED INTO BTC & ETH… SO WHY IS PRICE STILL STUCK? 👀💰
This might be one of the biggest contradictions in crypto right now.
Institutional demand is clearly improving.
But price action?
Still hesitant.
The latest weekly ETF numbers:
🟠 $BTC: ~$853.5M
🔵 $ETH: ~$244.9M
That’s roughly $1.1B flowing into BTC and ETH combined.
And yet $BTC is still stuck around the mid-$60K region instead of breaking higher with conviction.
So what’s going on?
There are a few possibilities.
🏦 ETF demand is being absorbed by existing sellers.
📉 Traders may be taking profits into resistance.
⚠️ Derivatives leverage could also be overpowering spot demand in the short term.
That’s why I wouldn’t look at ETF flows in isolation.
The bigger question is:
What happens if these inflows keep coming?
Imagine this:
🏦 ETF buying continues
📉 Selling pressure fades
🇺🇸 CPI comes in favorably
💧 Liquidity improves
At some point, available supply starts getting thinner.
And when that happens, a market that has looked completely stuck can move very quickly. 👀
But there’s another side to this.
If ETF inflows start weakening while $BTC keeps getting rejected at resistance, the market could be telling us that institutional demand still isn’t strong enough to overpower distribution.
That’s why I’m watching flow persistence, not just one impressive weekly number.
One strong week can change sentiment.
Several consecutive weeks can change the market structure.
👀 $1.1B has arrived.
Now the real question is:
Can it actually move the market?
#BTC #ETH #Bitcoin #Ethereum #ETF #Institutional #Crypto #Liquidity #AIInfraEarningsWatch📊 $SNDK Contract Liquidation Express (August 16)
According to liquidation data, the bulls and bears repeatedly squeeze and squeeze, while the dog farmer buys back and sells...
Time: Total liquidation, long liquidation, short liquidation
1 hour $4,044.48 $4,044.48 $0
4 hours $37,000 $6,961.45 $30,000
12 hours: $487,300, $273,400, $214,000
24 hours: $3.3067 million, $1.9221 million, $1.3846 million
From $SNDK's liquidation data, 1-hour long liquidations crushed shorts, with zero bears, a flash start with long selling; 4-hour direction suddenly reversed, short liquidation crushed bulls, shorts were 4.3 times longer, and short squeezes surged; 12-hour bulls overtook again, bulls 1.28 times longer, making a comeback; 24-hour bullish advantage persisted but narrowed sharply, dropping to 1.39 times, with long-term bear resistance surging — short liquidations soared from zero in 1 hour to $1.38 million. Dog Zhuang completed repeated bull and short squeeze on SNDK — selling long → short squeezing→ then → long and short tug-of-war, cumulative liquidation exceeded $3.3 million. Everyone control their positions carefully to avoid being bought back.
🔥 Market Barometer | August 16
Today's three hot topics point to the same theme: the capital feast of AI infrastructure is facing a brutal test from "burning money" to "making money."
🏗️ AI infrastructure financial reports follow up: the market only values "real money"
During Q2 earnings season, Wall Street's logic has completely shifted. Amazon AWS's revenue surged 37% year-on-year, Microsoft Azure soared 43%, and the three major cloud businesses grew 48% combined. What truly propelled Amazon into the $3 trillion market cap club was AWS's highest growth rate in 18 quarters.
However, the market is not "buying AI immediately." Meta delivered a better-than-expected earnings report but fell in after-hours hours because AI investment had yet to form an independent revenue stream; Nvidia rose only 2% for the week, with investors holding their breath awaiting the August 26 earnings report. The market's reward is no longer "who invests more," but "who makes money fast."
📊 CPI released tonight: The scale for a rate hike in September hangs in the balance
At 20:30 Beijing time on August 12, the US July CPI will be released. The Cleveland Fed forecasts that overall CPI for July will rise only slightly by 0.09% month-on-month, but core CPI is expected to rise 0.21% month-on-month, showing a rebound from June's flat month-on-month figure.
Currently, the market prices in a rate hike in September at about 44%-55%. If tonight's data exceeds expectations, the hawkish camp will quickly expand; If moderate, rate hike expectations may fade further. This data will be the first domino to determine the direction of the September FOMC meeting.
💰 NVIDIA drives 500 billion yuan in AI infrastructure financing: GPUs become "investable assets"
On August 10, NVIDIA announced cooperation with six giants—Apollo, BlackRock, BlackRock, Goldman Sachs, and KKR—to establish an independent computing power financing platform, aiming to leverage over $500 billion in third-party capital. Jensen Huang personally visited Wall Street, with all six institutions present.
However, on the day the news was released, Nvidia's stock price instead fell by about 2.8%, wiping out over $70 billion in market value. Michael Burry, the prototype behind "The Big Short," publicly warned that the "circular financing" model could repeat the borrowing chaos before the bursting of the internet bubble in 2000. Jensen Huang emphasized that AI computing power is already "a new era essential infrastructure equivalent to electricity and the internet."
💎 Summary
The earnings season for AI infrastructure proves one thing: the market no longer pays for "stories" but only prices "returns." NVIDIA's 500 billion yuan financing plan is the climax—and the biggest bet—in this capital game. And every basis point of tonight's CPI could determine the macro tone of this gamble. #财报观察员: AI infrastructure earnings report relay debut
#本周三CPI公布, will the pricing for a rate hike in September be rewritten?
#英伟达推动5000亿美元AI基建融资 The sense of fragmentation in the internal rotation of U.S. tech companies is truly at its peak, with no coherence in cyclical transitions.
Previously, the industry was heavily betting on the storage cycle, assuming that the tight memory supply and demand situation could hold out until 2028. Many people simply took long-term logic and took heavy positions to put them down. As a result, throughout July, the entire storage sector saw concentrated capital liquidation, and individual stocks collectively pulled back deeply.
In less than thirty days, funds have directly switched gears, and the market has uniformly shifted to a main theme of optical modules strengthening and storage under pressure. This market style changes at the slightest momentum makes it difficult to achieve stable returns with long-term narratives. Most of the time, you can only focus on market technical patterns and the flow of major funds for short-term swing trading.
▶️ Massive capital is clustering in the tech sector
1. This year, global technology-themed funds have accumulated a net inflow of $131 billion, directly surpassing the historical peak for the entire year of 2025, with an increase of $50 billion;
2. The annual inflow scale of technology funds has been steadily increasing for four consecutive years. Based on current inflow rates, the total inflow for technology funds in 2026 is expected to reach $216 billion, setting a new record;
3. Last week alone, $9.6 billion flowed in, bringing total inflows into US stock equity funds for the year to reach $652 billion—the highest annual inflow ever.
On one hand, massive amounts of hot money are flooding into major tech sectors, pushing up overall valuations; on the other, there is severe capital divergence across the AI computing power industry chain, with margin for error being infinitely squeezed, making the trading environment increasingly difficult.
$AAOI $QQQ 比特币在64,000美元附近反复挣扎,一个关键的链上指标刚刚亮起了黄灯——但不是红灯。 交易员Murphy发文指出,Glassnode的BTC卖方衰竭指数已在本轮熊市中首次进入“极端衰竭区”,该指标同时衡量低波动与高亏损状态。简单说,市场已进入“想卖的人都卖得差不多了”的阶段。 历史上,类似信号通常出现在熊市底部区间。2015年、2018年、2022年都曾出现过类似情况。 但Murphy特别强调了一点:首次触发并不一定对应绝对低点。有些周期甚至触发过多次,第一次进“极端区”之后,价格可能还会再跌一段。若后续价格维持震荡或继续走低,而指数不再创新低,则第二次信号的历史确定性通常更高。 当卖方衰竭信号与近期ETF连续流出、CME机构头寸转变以及价格在64K附近的反复试探叠加在一起时,市场的确是“卖方快没力气了”——但“没力气”不等同于“已经见底”,更不等同于“马上反弹”。Murphy的结论是:已经建仓的投资者并非错误,等待更明确信号再布局也可以,但若后续极端信号出现后仍不敢买入,可能错过后续行情。 $BTC $ETH $BTC #本周三CPI公布,9月加息定价会改写吗? #现货ETF资🦅OKX Zero-Downtime Cross-Border Migration Trading System Interpretation
Transactions across the entire network were uninterrupted, and OKX completed the cross-border migration of its core trading structure
OKX CEO Star publicly disclosed that the platform completed the cross-border migration of its core trading system last week. During the migration phase, there were no shutdown or crashes on the user side; in just half an hour to forty minutes, the response speed to trading instructions slowed down, but trading operations remained smooth throughout.
🐍 Analysis of the difficulty of migration technology
The cryptocurrency market operates continuously around the clock, with no system maintenance gaps after traditional securities close, which significantly raises the threshold for online migration.
The relocation includes not only hardware servers and underlying code, but also user order information, account asset limits, real-time market trends, risk control mechanisms, and API integration links all need to be migrated simultaneously. If any data synchronization error occurs at any stage, it can easily cause duplicate order submissions, abnormal asset amounts display, or even false triggers of forced contract liquidation, leaving very little margin for error.
This time, only minor delays and uninterrupted transactions were completed, making it the first of its kind among exchanges of similar scale in the industry and demonstrating its technological reserves.
🐢 Events hide signals
This announcement does not disclose the countries or regions corresponding to the migration, nor does it explain business adjustments or compliance supervision or other reasons for the move, so there is no need for blind speculation or interpretation.
This incident essentially reflects that OKX is building a multi-regional distributed operation and maintenance framework, strengthening emergency switching capabilities for remote disasters, and improving the overall risk resilience of its infrastructure.
🐎 Deeper view
Judging a exchange's technical strength and whether it can handle massive traffic during a surge is just the basic criterion;
Being able to complete system iterations and cross-border architecture migrations without server downtime is the core competitiveness for long-term stable platform operation $OKB $BEAT BTC 一小時提及 50 次,討論速度仍要放回全天看
OKX Onchain OS 在 08 月 11 日 11:00 記錄到 BTC 一小時 50 次提及,其中 X 49 次、新聞 1 次。
和二十四小時每小時平均相比,這一輪速度是 0.75 倍,屬於「有所放慢」;語氣則是偏多 34%、偏空 18%。兩條線沒有必要硬湊成同一個結論:熱度回答多少人在談,語氣回答文本偏向哪邊,兩者都不能直接代替成交與資金流。
下一輪若速度、新聞來源和實際市場成交一起延續,再提高判斷信心;若很快回到均值,這次變化就更像短窗噪音。Supposedly,
Fort Knox holds 147m oz of gold.
Nobody has fully audited it since the 1950s.
How long would it actually take?
147M oz ÷ 400 oz per bar = 368k bars
Each bar needs 3 things
1. Weigh it
2. Ultrasound it (tungsten fakes weigh almost exactly the same)
3. Log it
3min per bar.
About 18.4k hours of total work.
Bitcoin audits its entire coin supply with one command under 60 seconds.
Don't trust. Verify.#AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra 🦅 Tomorrow night, the U.S. CPI data will be released, which may rewrite expectations for a Fed rate hike in September
On Wednesday evening, the US July CPI inflation data will be released, which will directly determine whether the Fed will raise interest rates in September. Currently, the entire market is in a wait-and-see mode, with the whole market holding its breath and waiting for the results.
🐢 Current Interest Rate Pricing Status
The CME Fed Interest Rate Observation Tool shows: the probability of keeping rates unchanged in September is 55.6%, and the probability of a 25 basis point hike is 44.4%, with the two rates nearly evenly split.
If the data deviates even slightly from expectations, the market's pricing logic for rate hikes will quickly reset, amplifying market volatility.
Inflation expectations reference
Overall CPI year-on-year forecast is 3.4%, compared to the previous value of 3.5%; Core CPI is expected to be 2.5%, indicating that inflation is steadily declining.
But uncertainties lurk: in July, escalating geopolitical tensions in the Middle East, oil prices returning to the $80 mark, a rebound in energy sectors will drive up inflation, and CPI data risks falling short of expectations.
🦊 Three types of market scenario scenarios
1. CPI Below Expectations (Bullish)
Expectations for rate hikes cooled sharply, the US dollar weakened and declined, BTC and ETH led the rally, and all risk assets saw a broad rally.
2. CPI meets expectations (neutral)
Both positive and negative market factors failed to be realized, the market continued to fluctuate within a narrow range, and funds remained cautious awaiting further guidance.
3. CPI Rises Higher Than Expected (Bearish)
The probability of a rate hike has surged to 60%-70%, liquidity tightening expectations are suppressing the market, and cryptocurrencies are very likely to enter a round of downward correction.
🐻 Current market status
BTC fluctuated around 64,000, ETH held below 1,900, and SOL hovered around $76. The three major major coins have been declining for days, and before the data is released, funds are highly willing to avoid risks, mostly choosing to reduce positions and wait and see.
Key reference points for each currency
$BTC: 64,000 is the dividing line between bulls and bears; a valid breakout would target support at 60,000~61,000; If CPI positive news materializes, it could reach 68,000
$ETH: The trend has been weaker than Bitcoin's for a long time. 1850 serves as short-term support, and when negative news falls, the decline is often even greater
$SOL: The steepest decline, falling from the 294 high to 76, with 73 as a strong support level; Once positive news arrives, the oversold rebound will be much stronger than other coins
🐎 Operational Approach
Never heavily bet on one-sided positions before the data is released. Insertion and volatility around the CPI release can easily wipe out chasing positions. Wait for the data to come in and the market direction is clear before following the trend. There's no need to force yourself to capture the entire market period.
By the way, are you optimistic about this CPI data showing an unexpectedly unexpectedly unpredictable trend? $BEAT $ETH $SOL $BTC$BICO$OKB #本周三CPI公布, will the September rate hike pricing be rewritten? #Strategy再卖1690枚BTC, corporate finances are showing divergence #英伟达推动5000亿美元AI基建融资 The next AI infrastructure test is less about headline demand than where that demand converts into durable economics. Lumentum, CoreWeave, Coherent, Applied Materials and Cisco span optics, cloud capacity, chip equipment and networking, giving investors several views of the same spending cycle.
My read: orders alone may not settle the debate. Revenue conversion and margin support will matter more, because they reveal whether capacity investment is translating into pricing power or merely higher volume. SpaceX’s Aug 20 eligibility event adds a separate liquidity check, with another ~7% of locked shares potentially entering the market. Not advice, just analysis.
#AIInfraEarningsWatchThe 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#AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra 📌 Core conclusions from crypto afternoon
• Current price: BTC ≈ $63,990 (holding the 63,800–64,000 watershed); ETH ≈ $1,875 (losing 1,900, supporting 1,865–1,870, holding below 1,900→1,925).
• Market Front: Hormuz deadlock + oil price rebound + Strategy reduced holdings by 1,690 BTC, dual-currency gave back early gains and weakened, Alt followed the decline but did not rise (SOL -1.5%, ADA -5.2%, BNB only -0.7%), fear and greed at 29.
• CPI Two scenarios (Beijing time 8/12 20:30 US July CPI): Below expectations (head-to-head <average 3.4% / core < 2.5%)→ Rate cut expectations rebound, BTC tests 65k, ETH rebounds to 1,900+; Above expectations → tightening concerns, BTC tests 63.3k, ETH tests 1,840–1,850. 很多人现在只盯着: “BTC什么时候突破?” 但我更关注另一个问题: 👉 资金正在提前布局什么? 目前 $BTC 仍在 $64K 附近震荡,重新站稳 $65K 之前,市场情绪依然偏谨慎。 而真正值得注意的是: 美国现货 BTC ETF 最近连续几个交易日保持资金流入,单看最新几轮交易日,累计流入已经达到数亿美元级别。 与此同时,ETH ETF 也持续获得资金关注。 但奇怪的是—— 💰 资金在进场 📉 价格却没有同步爆发 这恰恰是我现在最关注的市场信号。 因为 ETF 资金流和散户风险偏好并不一定同时变化。 机构可能正在吸收流动性,而市场却还没有真正进入全面 Risk-On。 这意味着: 真正的轮动,可能还没有开始。 --- 👑 $BTC — 市场总开关 BTC 目前依然是整个市场最重要的方向确认。 我会重点观察: 🔹 $62K–$63K 支撑 🔹 $65K 能否重新站稳 🔹 $67K 附近是否出现突破量能 如果 BTC 能够守住关键支撑,同时重新突破 $65K,市场风险偏好可能进一步改善。 但如果再次跌破关键支撑,那么所谓的“资金轮动”仍然可能只是短期假象。 --- 🏛Tens of millions of dollars in short positions emerge! A whale went all out on SKHX with 4x leverage in the perpetual contract market, with a position value as high as $20.18 million, an average entry price of about $1,010, a current unrealized profit of $23,000, and total profit and loss over the past week reaching $1.01 million.
Short positions occupy 100% of the position, with margin utilization reaching 100.44%, nearly fully invested. If the price continues to fall, the whale will gain huge profits; Conversely, if the price rebounds to near the liquidation line of $1931, there is a risk of full margin liquidation, and under a bearish pattern, close attention should be paid to price movements $SKHYNIX In fact, the arrival of a bull market is always recognized belatedly.
The real market turning points often occur in the quietest moments. When most people are still debating "whether the market will have another round of decline," a few funds have already quietly started adjusting their positions. The current crypto market is at such a delicate stage—stabilization signals are accumulating but have not yet been fully priced in.
1. The market is stabilizing, and signals are gradually becoming clear
First, there are currently no obvious systemic negative factors in the market. On the macro level, the Federal Reserve's expected path is relatively stable, with no sudden tightening shocks; on the regulatory side, no major negative policies have been implemented; on the funding side, spot ETFs continue to maintain net inflows, and institutional funds have not massively withdrawn. Although risk appetite remains cautious, panic sentiment has clearly cooled down, which is an important premise for stabilization.
Second, the structure of mainstream assets is improving. Both BTC and ETH have not hit new lows for this phase, and their prices repeatedly find support in key zones. BTC is consolidating around the $65,000 level, while ETH is stabilizing near $1,900. Both show characteristics of "no longer breaking downwards," and the previous high areas above still hold potential for breakthroughs. Historical experience shows that when mainstream assets hold their previous lows and gradually repair moving average structures, it often signals the prelude to a mid-term market restart.
From a technical perspective, daily-level MA and EMA are being repaired. Short-term moving averages are beginning to flatten or even form golden crosses. If prices can continue to hold steady and drive the 60-day and 120-day mid-to-long-term moving averages from downward trends to flat or even upward trends, the market structure will significantly improve. Once this process is complete, the market is more likely to welcome a mid-to-small scale upward trend supported by volume.
These changes are not drastic but are enough for prescient funds to position themselves in advance.
2. After market stabilization, alpha targets deserve more attention
When the market shifts from decline or sideways movement to mild upward movement, funds usually do not just stay in BTC and ETH. Targets with independent narratives, high elasticity, or ecosystem binding are more likely to generate excess returns. Currently, three categories deserve focus: $OKB, $CRCL, $SOL.
OKB's excess logic comes from ecosystem binding and scarcity. It has evolved from a simple platform token to the only native Gas token of the X Layer, required for on-chain transfers and contract interactions. Coupled with a permanently capped supply of 21 million tokens and a closed-loop layout of exchange + wallet + public chain, OKB is more likely to benefit simultaneously from increased trading activity and on-chain demand when the market warms up, often showing higher elasticity than the overall market.
CRCL (Circle) represents the stablecoin and compliance infrastructure sector. As the issuer of USDC, its value is highly correlated with stablecoin circulation, on-chain adoption, and institutional inflow demand. When market risk appetite rises and on-chain activity increases, stablecoin demand usually expands in tandem. CRCL combines US stock attributes with exposure to the core crypto infrastructure sector, making it easy to receive a revaluation of "infrastructure" by funds during an upward market phase.
SOL is a typical high-alpha representative. The Solana ecosystem still holds advantages in transaction speed, developer activity, and application deployment. Historical data shows that during BTC-led early or mid-stage rallies, L1 assets like SOL often experience significant amplified gains. Once the market confirms stabilization and releases liquidity, high-elasticity public chains are more likely to become targets for capital pursuit.
The common feature of these three categories is that they do not just "follow the market up" but each is tied to a clear independent narrative—ecosystem consumption, compliance infrastructure, and high-performance public chains. When the market shifts from defense to offense, funds will prioritize these directions that can provide alpha.You can't make that money because you make decisions for the market 🤷
BTC broke 64,000, ETH broke 1,900. I held short positions for days without touching, and once I left, it was like a waterfall. In that moment, my heart wasn't anger, but a calm 'just as expected.'
When you take short positions, the market doesn't move; once you move the market, it starts—not monitoring, not targeting, but making decisions for the market.
"It should have fallen, right?" "Pretty much now, right?" "If it doesn't fall soon, will it reverse?" Once these thoughts arise, your position is close to being closed. Because you start using "I think" instead of "market talk." You judge the right direction, but you exit early before the market gives a confirmation signal. The market isn't targeting you; you are ahead of the signal.
Many people look in the right direction but end up not making money. It's not because the direction is wrong, but because the day the direction is realized never comes. Many people look in the right direction, but few can hold the order and wait for the direction to come out. It's not about ability, but about patience.
But patience can't be quickly made; you have to slowly build it through repeated "falling right after you leave" experience. But if you don't reflect on that "why" after each missed opportunity, the same storyline will keep repeating.
If you miss this wave, just miss it. As long as you're still at this table, the next opportunity will come sooner or later. But next time, don't make decisions for the market again—let the market tell you where it's going, then follow it, don't jump ahead.
#BTC #ETH #SNDK #交易心理$BTC $ETH $GRVT #财报观察员: AI infrastructure earnings debut 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 Supposedly,
Fort Knox holds 147m oz of gold.
Nobody has fully audited it since the 1950s.
How long would it actually take?
147M oz ÷ 400 oz per bar = 368k bars
Each bar needs 3 things
1. Weigh it
2. Ultrasound it (tungsten fakes weigh almost exactly the same)
3. Log it
3min per bar.
About 18.4k hours of total work.
Bitcoin audits its entire coin supply with one command under 60 seconds.
Don't trust. Verify.#AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra Bank of Korea Senior Deputy Governor: Additional rate hikes expected!!
Background Analysis: Why Such a Firm Stance?
This hawkish statement is not an isolated incident but is built on a series of recent economic data and policy groundwork:
The rate hike cycle has long begun: On July 16, 2026, the Bank of Korea announced a 25 basis point rate hike to 2.75%, ending the previous easing cycle and marking a formal shift in monetary policy. This statement reaffirms and emphasizes that this tightening path will continue
Hawkish signals continue to be set: As early as May this year, the senior vice governor stated that given the better-than-expected economy, it was time to "stop cutting rates and start considering raising rates." The central bank's advisory board also sent hawkish signals in early August. These factors laid the groundwork for the statement of this "additional rate hike."
A strong economy "backs up" tightening: data shows that South Korea's GDP grew 3.7% year-on-year in the second quarter, and total domestic income surged 15.6%, reaching the highest level in nearly 38 years. Meanwhile, the broad money supply (M2) in May grew 11.7% year-on-year, the fastest growth since February 2022. Robust economic growth and rapidly expanding liquidity provide fundamental support for the Bank of Korea's continued rate hikes
Overall, the Bank of Korea's decision clearly shows that, in the face of strong economic data and persistent inflation and exchange rate pressures, its policy balance is now fully tilted toward further interest rate hikes to stabilize prices and exchange rates! #财报观察员: AI infrastructure earnings report debuts one after another 🇺🇸 US MACRO: Clarity Act passes Senate
There is a legislative victory that most investors are overlooking:
On 8/8, the US Senate just pushed the Clarity Act — the first comprehensive legal framework for crypto — through a crucial step, marking Trump's second major win after last year's stablecoin law.
Reduced legal risk means institutional money can flow in more easily;
The direct beneficiaries are well-compliant coins like $XRP , $ADA, $SOL and the RWA group $LINK, $ONDO.
But don't forget the other side: the Fed still holds interest rates at 3.50–3.75% with a strong USD — liquidity hasn't been loosened;
Also, $TRUMP Media just canceled the treasury deal with Crypto.com causing $CRO to plunge, and Warren is pressuring the SEC on Trump's memecoin.
In my opinion, the Clarity Act is a long-term catalyst, while the Fed is the short-term key — if Chairman Warsh cuts rates at the next meeting, $BTC could break out from the $64k range.
What do you think, will the Clarity Act pass the House after the August recess, and who benefits the most — $XRP, $ADA, or $SOL?
#BTCETHETFFlowsDiverge #StrategySellsBTCAgain #CPIToResetFedBets #AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra 🚨 THE MARKET LOOKS BORING… BUT THAT’S EXACTLY WHAT HAS ME PAYING ATTENTION. 👀💰
Everyone is waiting for the next big $BTC breakout.
I’m watching the money underneath the price.
$BTC is hovering around $64K after losing its $65K+ momentum, and traders are getting cautious ahead of U.S. July CPI.
Fear is still hanging around.
But here’s the interesting part:
Institutional demand hasn’t completely gone away.
U.S. spot $BTC ETFs reportedly pulled in around $853.5M over five straight sessions, while $ETH ETFs added roughly $244.9M during the same week.
Yet price still isn’t exploding.
Why?
Because institutional accumulation and retail risk appetite are two completely different things.
And that gap could become very important. 👀
👑 $BTC — THE GATEKEEPER
Bitcoin needs to stabilize first.
If $BTC can defend support and regain momentum, the entire crypto market gets room to breathe.
🏛️ $ETH — THE ROTATION SIGNAL
$2K is the level I’m watching.
A clean reclaim with volume would make the altcoin setup much more interesting.
⚡ $SOL — THE RISK GAUGE
$SOL continues to show relative strength.
If traders start feeling comfortable taking more risk again, SOL is one of the first charts I’ll be watching.
🔥 ALTCOIN RADAR
$SOL • $XRP • $HYPE • $SUI • $TAO • $WLD • $JTO • $ONDO • $AAVE
👀 EARLY ROTATION WATCH
$HUMA • $ZKP • $METIS • $EDEN • $MEME
But I’m not calling altseason yet.
One or two green candles don’t mean anything.
I want to see the whole picture line up:
✅ Breadth improving
✅ Volume returning
✅ Liquidity expanding
✅ BTC staying stable
✅ Capital actually rotating into alts
My checklist is simple:
CPI → BTC reaction → ETH strength → BTC dominance → Altcoin volume
If CPI improves risk sentiment and Bitcoin holds its ground
But sometimes the quietest part of the market is where positioning happens before everyone starts paying attention. 👀🔥
Which $ALT are you watching before the next rotation?
Drop it below. 👇
Market observations only. Not financial advice. DYOR.
#DailyOrbit #Crypto #Bitcoin #Ethereum #Solana #Altcoins #Altseason Something notable just happened on the Toronto Stock Exchange, and it says more about where institutional finance is heading than most headlines this month. On Monday, August 10, BlackRock's Canadian arm rolled out a new fund called IBQT — the iShares Equity + Bitcoin ETF Portfolio — built on a simple but telling formula: 97% traditional global stocks, 3% Bitcoin, wrapped into a single ticker. No need to buy a crypto fund and a stock fund separately and rebalance them yourself. BlackRock did theThe 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#AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra On August 10, TSMC announced its July monthly revenue, with monthly revenue reaching NT$467.58 billion, a year-on-year surge of 44.7% and a quarter-on-quarter increase of 5.6%; From January to July, cumulative revenue reached NT$2.87 trillion, a year-on-year increase of 37%, setting a new record for monthly revenue. July's data was significantly stronger than seasonal patterns. In previous years, month-on-month growth in July was only 1-2%, but this year's 5.6% increase reflects continued AI chip orders, with third-quarter results potentially surpassing the upper limit of guidance. Advanced process capacity remains full, with 3nm, 2nm, and CoWoS advanced packaging capacity in short supply. Leading clients like Nvidia, Google, and Apple continue to lock up capacity, and high-performance computing remains the primary growth engine. Impressive monthly data also brought market divergence. On one hand, data confirms that AI hardware demand is not just a hype, and computing power capital expenditure is real; On the other hand, the stock price has already fully priced in high growth, and combined with the massive annual capital expenditure of $60-64 billion, overseas factory construction and new production line ramp-up will exert dilution pressure on future gross margins. After the announcement of US ADR news, there was volatility, with institutions diverging: bulls believe Q3 earnings may once again exceed guidance; Cautious opinion tip: highly concentrated on major AI clients, if downstream capital expenditure contracts, revenue will be directly impacted. The market will focus on tracking revenue data for August and September to verify the quality of third-quarter results, while also observing the trend of gross margin changes under high capital expenditures. #财报观察员: AI infrastructure financial reports take the stage $TSM $BTC "Spot gold hits a 9-week high"$XAU Holding above 4300 directly reflects safe-haven buying driven by geopolitical risks.
Dabing 2Bing is fluctuating sideways, with an unclear direction. It may continue to decline in the later stages.
Short-term trend: Before tonight's U.S. economic data release, the market is highly likely to maintain the current pattern: gold (XAUT) is oscillating with a strong side, $BTC is consolidating within a range, and $ETH is relatively weak.
Key variable: Closely watch the ADP employment data at 20:15 tonight. If the data falls significantly short of expectations, it could reinforce expectations of an "economic slowdown→ Fed rate cut," which could theoretically benefit gold and crypto markets; if the data is strong, it could trigger a dollar rebound, putting pressure on risk assets (especially ETH).
Brothers, with tonight's ADP data, do you bet on gold or are you optimistic about a dollar rebound? #本周三CPI公布, will the September rate hike pricing be rewritten? #ADP就业降温, Fed policy divergence is intensifying Looking at these three hot topics together is more important than looking at any one of them alone!! The first was $NVDA, in collaboration with institutions such as Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR, to promote an AI infrastructure financing platform worth over $500 billion. It is important to note here that this is not "Nvidia directly spending $500 billion to buy stocks," but rather plans to leverage third-party capital through financing platforms for AI chips, data centers, power infrastructure, and other infrastructure projects. Nvidia disclosed it could provide up to about $125 billion in support for potential deals. Meanwhile, Wall Street has previously referred to AI capital spending as the "AI capex supercycle," indicating that AI infrastructure is gradually shifting from merely tech company capital expenditures to large-scale financing themes involving financial institutions. The second issue is the July US CPI, which the market is about to face. According to the latest schedule from the U.S. Bureau of Labor Statistics, the July CPI will be released at 8:30 ET on August 12, which is the evening of August 12 Beijing time; June CPI year-on-year has already reached 3.5%, and core CPI is 2.6% year-on-year, so the significance of this data is not just about whether inflation has risen or fallen, but will directly influence how the market reprices the Fed's policy path in September. The third thing is that US AI earnings reports and AI infrastructure logic are continuously being transmitted throughout the entire risk asset market热榜都在说: Robinhood“杀进英国加密市场”。$BTC 先纠正一下。$ETH 目前Robinhood英国App里, Crypto还没有正式开放。$GRVT 官方说的是: 很快上线。 但真正值得看的,不是晚几天还是早几天。 而是—— Robinhood正在把自己变成一个 “什么都能交易”的超级入口。 股票+期权+期货+ Crypto 全部塞进一个App。 而且就在昨天, Robinhood旗下Bitstamp 又把BTC、ETH交易带进了澳大利亚。 这说明它不是只盯英国。 而是在全球抢Crypto用户。 更有意思的是: Robinhood Q2加密交易收入 同比还下降了38%。 收入在降,它却继续扩张。 这才是重点。 它赌的不是下一根BTC阳线, 而是未来散户的交易入口。 一旦英国Crypto正式开放, 真正有压力的可能不只是本土券商, 传统加密交易平台也要重新抢用户。 —————— 📍BTC我还是按原来的低频计划: 现在约64.1K, 不追。 63100—63300:重点接多区 63180附近:核心观察 62700—62850:极限补仓区 62380下方:失效 目标: UBS calls for 1625, Morgan Stanley says SK Hynix can't rise that much! Micron and SK Hynix—who do you really trust?
One shouted, "It can still rise by 85%," while the other said, "The price increase isn't that much." Two top institutions face off from afar—who is lying?
UBS just released a report maintaining Micron's "buy" and a $1,625 target price, which is still 85% higher than the current stock price. The reason given is that HBM is still in short supply, and after Nvidia adjusts its allocation, total HBM consumption in 2027 will actually be higher, with DRAM shortages at least through 2028. UBS even raised Micron's 2028 EPS to $265, saying free cash flow could accumulate to $450 billion.
JPMorgan Chase has started pouring cold water — the market expects SK Hynix's HBM contract price to rise by more than 50% in 2027, but Morgan Stanley forecasts an increase of less than 40%. The reason is that Nvidia is too strong, and SK Hynix's ability to reprice annually is limited.
One focuses on the long term, the other on rhythm. UBS is betting on "long-term insufficient HBM," while Morgan Stanley is focused on "short-term prices are not optimistic." The fundamentals remain unchanged; what has changed is that market expectations are being repriced.
Old Zhang's view: HBM's "volume" is still rising, but "price" expectations are narrowing. The long-term logic of the storage sector remains unchanged, but short-term earnings reports will be more and more disturbed.
If you also play with US/Korean stock tokens, please follow Lao Zhang.
#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
$MU $SKHYNIX $CL #财报观察员: AI infrastructure earnings report debuts one after another
Many people only watch whether BTC can pull back, not that the US stock market just finished a "AI infrastructure earnings relay" in the past two weeks—
Google, Microsoft, Meta, and Amazon all hand over their papers in Q2 2026:
• The four CAPTEX companies combined for a single quarter of $171.2 billion, continuing to surge year-on-year
• Amazon's full-year guidance is revised up to $220 billion, Google $195–$205 billion, Meta $130–$145 billion
• Microsoft single-quarter capex 41 billion (+70% year-on-year), Azure +43% year-on-year
• Even more impressive are the orders on hand: the combined outstanding orders from four companies totaled about $2.33 trillion, up +188% year-on-year
Plain language translation: The market used to fear "AI burning money and going down the drain," but now it's become "orders piling up to the ceiling, but computing power isn't enough to sell." AI infrastructure is not a receding tide, but entering a positive cycle.
What does that have to do with the crypto world?
1. For AI narrative coins like TAO / FET / RNDR / GRT, NVDA + cloud factory earnings reports act as macro sentiment switches, with correlation of 40–60%.
2. Listed mining companies (IREN, TeraWulf, Riot) are subletting power and data centers to AI clients. Riot just signed a $9.1 billion long-term contract for computing power with Anthropic, signaling a revaluation of mining stocks
3. On the capital side, semiconductor ETFs attracted over 20B in capital in the first half of the year, while BTC ETFs saw net outflows during the same period—marginal US dollars were switching back and forth between AI and crypto, and strong AI earnings → risk appetite returned→ easing pressure on BTC
But don't get carried away: in 2027, the four major companies' CAEX growth rate is expected to slow down, and when expectations are "overhyped," good news turns negative (referencing AMD's 8% post-market drop in Q2).
My view this time: the AI infrastructure earnings season is meant to provide a bottom for crypto risk assets, not a direct pull-up button. Before BTC breaks out, AI sector coins are more comfortable than blindly chasing memes.这个热度很高的事件【光通信大战存储】好像很少看到简中区的朋友讨论。我简单介绍一下事件背景: @jukan05 因为头像是黑头发被很多人称为黑毛股神,之前一直是内存多头。但是这次宣布已经清仓内存股,转而做多光通信。比如 $AAOI 凭借亮眼的第二季度财报,股价从 70多刀 拉升到最高 140多刀。 @aleabitoreddit 白毛股神在简中推特的名气更大一些,其提出的“卡脖子”投资理论被很多人熟知。她认为存储的基本面并没变。光通信的暴涨只是回归正常估值,目前存储的市值是被低估的。
下面是我的理解:所以到底应该买存储,还是买光?我觉得没必要把它变成一道单选题。
“多光空存”更像一种交易策略,而不是产业趋势的最终答案。
光通信目前的优势,是景气度高、订单确定性强,而且1.6T、CPO等新技术仍然在持续推进。
但它的问题也很明确:随着模块逐渐标准化、大规模自动化生产,制造壁垒可能下降。
存储则完全是另一套逻辑。现在它正在经历一次非常痛苦的预期调整:大家开始担心价格见顶、产能扩张、利润增速下降。这些担忧并非没有道理。比如最近大家能看到 $MU $SKHY 股价承压。
但另一方面Tuesday, August 11 — the crypto market is holding its breath
Bitcoin is parked at $63,989, Ethereum at $1,873, and $XRP sitting at $1.01 — a market that's gone quiet rather than volatile, which in crypto usually means everyone's waiting on the same catalyst.
That catalyst is September. The Senate's procedural move on the crypto market-structure bill last week didn't put anything into law — it just cleared the runway. The actual floor vote lands when lawmakers return from recess, Sept 14–16, and it still needs several Democrats to cross over on unresolved ethics language. Until then, this is a market pricing in a maybe, not a done deal.
Underneath the calm, two stories are worth tracking:
Security is having a rough week. Payment processor Coinsbuy got drained of roughly $8 million across Tron and Ethereum in a coordinated attack — funds were laundered through instant-exchange services before some got frozen.
Meanwhile, OpenAI made an unusual disclosure: its next model, Astra, showed cyber capabilities strong enough that the company can't rule out it hitting the highest risk tier in its own safety framework. Development isn't stopped, but it's now boxed into isolated testing with government and safety-org oversight. Two different corners of the tech world, same underlying theme — capability is outrunning containment.
The builders keep building anyway. Vitalik Buterin's latest roadmap update leans into quantum resistance, privacy, and AI-assisted security — a signal that Ethereum's core team is thinking in years, not news cycles.
Net read: the macro setup (a possible September regulatory unlock, a Fed still sitting tight at elevated rates) is doing more to shape sentiment right now than any single day's price move. Worth watching the calendar more than the chart this week.
#AIInfraEarningsWatch #CPIToResetFedBets #Nvidia500BAIInfra $BTC $ETH $XRP. Sources
CoinDesk — "Crypto exchange Coinsbuy loses $8 million in coordinated two-blockchain attack" (Aug 10, 2026)
OpenAI (official) — "Responding to the next frontier of critical cyber capabilities" — openai.com