Orbit Post Sitemap

I just sat down and glanced at last night's US stock market close: the Dow fell 0.04% to 53,770 points, the S&P 500 rose 0.26% to 7,748 points, and the Nasdaq gained 0.54% to 26,588 points. On the surface, everything was calm, but inside, the market was sharply divided. CPI was implemented, but the market gave no respect US July CPI was 3.4% year-on-year, core 2.5%, and 0.1% month-on-month—four figures perfectly matching expectations. After the data was released, the probability of a rate hike in September dropped from 47% to 45%, but that's about it. Meeting expectations is the biggest problem in itself—the market had already priced in the expectation of cooling inflation two weeks in advance, but when the data actually came out, it turned out to be as expected, and there was no buyer. AI infrastructure earnings reports exploded collectively—this is the real highlight of tonight CoreWeave's Q2 revenue was $2.575 billion, a year-on-year surge of 112%, exceeding expectations. Its stock price jumped 19%. Lumentum was even stronger, doubling year-on-year to $1.01 billion in Q4 revenue, with its stock price up over 13%. Coherent's Q4 revenue was $2.05 billion, up 34% year-on-year, but it actually fell in after-hours trading. Performance exceeded expectations and even dropped; market expectations have already been pushed to the ceiling by AI—it's normal to exceed expectations, but even the slightest disappointment is a capital offense. SPCX rose another 9%, and Musk said internally that AI revenue would surpass all other revenues in September, and that in five years, AI will account for 99% of SpaceX's value. This is a big picture—the market is really buying it. The seven tech giants continued to be slashed. Meta fell over 3%, Microsoft over 2%, Amazon Tesla dropped over 1%, and Apple and Google edged down. Nvidia, however, rose 3%. Memory chips surged across the board. SK Hynix rose over 9%, Seagate gained over 7%, SanDisk rose over 5%, and Micron rose nearly 5%. Chinese concept stocks are in dire strait. The Nasdaq Golden Dragon China Index fell 2.37%, WeRideStock dropped over 9%, and BOSS Zhipin dropped nearly 4%. For BTC, CPI has landed but no direction is given. BTC is still hovering around 64,000; BTC didn't follow US stocks rising, and BTC didn't follow US stocks as they fall. AI infrastructure surges, tech giants get slashed, Chinese concept stocks plunge—each going their own way in the same market, showing funds are readjusting, not fleeing. Tonight there's also PPI data, with market expectations of 0.3% month-on-month and an annual rate of around 5.5%. If PPI exceeds expectations, inflation worries will return. I'm not heavily positioned; I'll wait for PPI to be realized. Acting now is just gambling; there's no need. $SPCX $QQQ $SKHY 在夜深人静、烟雾缭绕的键盘前,看盘多年的人总会有一种直觉:市场从不相信永动机,但资本永远在疯狂寻找下一个物理瓶颈。今天看到 Lumentum 刚拉出来的这份成绩单——FY2026 Q4 营收直接飙升至 10.1 亿美元,同比暴增 109%,调整后 EPS 飙到了 3.23 美元,连带下一季度的业绩指引都敢直接指向上限 12.75 亿美元。老实说,这不仅仅是一份亮眼的数据,更是一记沉重响亮的警钟。 当所有人还在把目光死死盯在 GPU 显存和晶圆代工上的时候,真正经历过牛熊洗礼的老手早已看明白:AI 算力集群规模每翻一倍,计算的肿瘤就会向高速光互连(optical interconnects)扩散一分。光模块与激光器,早已从曾经的配套角色,变成了当今庞大 AI 算力帝国里最昂贵、最不可或缺的通道收费站。 但问题在于,这究竟是一场可持续数年的产业长牛,还是一场因短期集中采购而引发的产能幻觉?管理层口中“AI与云数据中心的强劲需求”,固然吹响了冲锋号,但半导体产业几十年来血淋淋的周期教训告诉我们,伴随着疯狂扩产而来的,往往也是随之而来的周期性剧烈震荡。资本是极其残酷的,今天能把你捧上神坛,明天$SPCX A strong comeback near $150? Rocket launch? Don't be blinded by your eyes and get too excited too soon! Here is an analysis of the Rockets' market situation and insights: Due to the low expectations for CPI data release and the temporary hype of negative news being realized after unlocking, the short-term upward momentum sentiment has been increased and restored. However, it seems that retail investors and those chasing the gains have forgotten that the overall market in July was in a slump, slump and slump. Although the data release boosted tokens in the AI and tech sectors, SPCX still faces risks of short-term pullbacks and high-level sell-offs. The signals of bullish and negative factors are very clear; short-term corrections are essential. After the phased hype and data turmoil, the rocket will have to fall back to the bottom. Therefore, I personally support the Air Force and will monitor the Rockets' market movements during the day. If you lack experience or knowledge of US stock sector tokens, you can exchange ideas and learn from him. $ETH $BTC Money is shifting internally: ETF outflows ≠ end of bull market, but a reshuffle before rotation On the surface, BTC is consolidating around 63800, ETH oscillates around 1895 for a day, SOL shows small fluctuations, and the market is lifeless. In essence: old hotspots cool down, new narratives accumulate, funds are switching between highs and lows, not exiting. Supporting signals: BTC spot ETF had a net outflow of about $144.6 million on 8/10, breaking the consecutive inflows, but on 8/12 IBIT saw a return inflow of $50.2 million, indicating institutional portfolio adjustment rather than a collapse retreat; On-chain whale transfer-to-exchange ratio rebounded, indicating swing reduction rather than bottom liquidation; The stalemate at the Hormuz Strait (Iran blocking passage, 8/18 negotiation window countdown) supports safe-haven demand, but gold itself is stagnating, not yet reaching global resonance speculation. Conclusion: liquidity tightening + internal rotation coexist, before direction is chosen, it’s all noise. ------ BTC trend framework (weekly → daily → operation) Weekly: weak rebound after second test, bear tail still needs one more shakeout Pattern: natural rebound after panic sell-off, low volume, weak supply and demand — typical mid-bottom formation feature, not a reversal. Judgment: the bear market end still lacks the final forced shakeout step, target directly smashing below 57k (coinciding with 0.618 Fibonacci 57825 and previous cycle support). Weekly action: short on rallies, do not bottom-fish. Daily: liquidity not fully taken, two paths await confirmation Ideal path: rebound to 67200 to sweep stop losses/take liquidity → weekly resonance short setup. Right-side path: directly break below 62500 → chase shorts, no bottom guessing. Reflection and adjustment: the weak rebound after second test should not consume chips to "must take liquidity," avoid short-term uncertain moves; after BTC long position profit-taking, no more short-term mainstream coin trades within the week. 3. Current anchor point (8/13) Current price ~63800, between 62500 support and 67200 liquidity overhead; ETF outflows + whale transfers = rebound lacks sustained buying, weak sideways likely to break down; Trigger right side: hourly/daily close below 62500 to short; or rebound to 67200 with stagnation to short structurally. ------ Gold XAUUSD framework Weekly Follow-up rebound after panic sell-off, with a second bottom test to come → short on rallies. Daily Currently 4360–4400 stagnation, 4400 is psychological + 100-day SMA resistance, touched 4435 on 8/11 then pulled back; Key support 4360: break leads to 4200; if 4200 holds → buy on dips logic valid; If retest 4360 sideways then rally → follow weekly short path. Drivers: Hormuz Strait unopened + Iran tough stance = safe-haven support, but CPI cooling + negative nonfarm payrolls shake rate hike expectations, lacking the "global resonance speculation" leg. ------ Position discipline in rotation market Altcoins only focus on leaders + those with clear logic; blacklist concept pumpers. Leave room in positions, don’t bet all on a single hotspot; rotation is zero-sum, first mover eats the latecomer. Avoid contracts: rotation market is sharp, leverage sweeps principal twice to the bottom. More important than "guessing the next leader" is managing principal; direction can be watched slowly, position size is the survival line. ------ Summary in one sentence BTC weekly waits for shakeout below 57k to short on rallies, daily waits for either break below 62.5k or rebound to 67.2k for confirmation; gold waits for 4360/4200 two-level support to define; macro money is shifting internally, rotation market is not about guts but clean position sizing. The above is technical analysis, not investment advice; contracts should strictly control position size and stop loss. $BTC $ETH Anthropic’s $3T Valuation: AI Breakthrough or VC Exit Strategy? Anthropic’s potential IPO valuation is reportedly becoming enormous, with expectations of $2T+ and possibly up to $3T later this year. That’s a massive jump from its roughly $965B valuation during its previous funding round. In just a few months, the company’s valuation could potentially multiply several times. Some see this as proof of the AI boom. But another interpretation is that VCs may be trying to exit while AI valuations and market liquidity remain extremely high. The valuation logic is straightforward: if Anthropic reaches $100–120B in annualized revenue, a 30× price-to-sales multiple would imply around $3T. The problem is whether a model company deserves the same premium as a company like NVIDIA. NVIDIA benefits from strong hardware demand, high margins and a powerful competitive position, while AI model providers face huge computing costs, open-source competition and rapidly falling API prices. So why rush toward an IPO? Because the exit window may be narrowing. Training costs for next-generation models keep rising, while performance improvements are becoming harder to achieve. VCs may want to sell into strong public-market liquidity before AI enthusiasm cools. This could also be a warning for crypto AI and DePIN projects. If even major Web2 AI companies are trying to monetize sky-high valuations through public markets, AI-related Web3 projects could face serious valuation and liquidity pressure once the market starts demanding real cash flow instead of narratives. For retail investors, blindly chasing AI stocks or AI tokens at these valuations could mean becoming the final buyers in the VC exit cycle. The big question is: if Anthropic actually goes public above $2T, will that valuation survive through explosive AI adoption—or could it face a major correction once the hype fades? #CPIEasesHikeBets #AIInfraEarningsWatch #KoreaChipsLeadRebound 🚨 X Synchronized Update | OCC Ushers in the Era of Crypto Banking: Crypto is Entering the U.S. Financial System 🇺🇸 The Office of the Comptroller of the Currency (OCC) is further opening the door for eligible digital asset companies to enter the U.S. national banking system. First, to correct a commonly misunderstood statement: This is not about "the U.S. approving all crypto companies to become national banks," but rather that regulation is establishing a clearer path—eligible digital asset companies can apply for national bank or national trust bank licenses and conduct corresponding operations after meeting regulatory requirements. What truly deserves attention is not the "additional bank licenses," but a deeper change: Crypto is gradually moving from outside the traditional banking system into the financial system. The past path is: Traditional finance → Crypto Banks, funds, ETFs, and asset management institutions are entering crypto. Now, another path begins to appear: Crypto → Bank Digital asset custodians, stablecoin companies, and crypto financial enterprises have begun to enter the banking system. Why is this important? Because what Crypto lacked in the past was not just assets, but complete financial infrastructure: Custody, payment, settlement, lending, asset management, fiat entry, and institutional services. And the banking system happens to provide these capabilities. This means Bitcoin's role may also change. Past: Bitcoin = Digital Gold In the future, it may further become: Stored-value assets + collateral + yield assets + payment assets + core financial infrastructure assets Bitcoin ETFs address: "How does traditional finance hold BTC?" BTCFi addresses: "What else can BTC do besides holding?" Bitcoin Banking may address the following issues: "How can ordinary people use BTC like they use a bank account?" This is also why projects like Bitcoin Neobank, BTCFi, and SatPay are starting to attract attention. If regulatory trends continue, the core of future competition in the crypto industry may no longer be just TVL, trading volume, and token market capitalization, but will gradually shift: Bank licenses, custody capabilities, payment capabilities, stablecoin capabilities, institutional clients, and financial infrastructure. So, OCC may be opening more than just a door for a "crypto bank." Instead, it is a new industrial pathway: Bitcoin → BTCFi → Banking → Payment → Everyday Finance Crypto has tried to build its own financial system in the past. In the future, what is more likely to happen is: Crypto is gradually becoming part of the traditional financial system. And this may be the next stage of Bitcoin's financialization truly worth paying attention to. $BTC $COREGOLD IS BACK ABOVE $4,400 — THIS IS NOT JUST A PUMP There is one thing I think the market is underestimating: Gold is returning to the $4,400/oz area while the biggest drivers of the bullish cycle have not gone away. On August 12, Gold Spot rose to about $4,406/oz, the highest level in more than 2 months. The direct driver came from the US CPI in July only increased by 0.1% compared to the previous month, causing the market to reduce expectations that the Fed will continue to raise interest rates. (Reuters) On August 13, gold corrected to around $4,374, but since the beginning of the month alone, the price has increased by more than 8%. (Reuters) In my opinion, to understand GOLD at this time, you have to look at 3 major cash flows. 1. INTEREST RATE – USD – REAL YIELD Gold does not generate yields. So when the market expects higher interest rates → USD and more attractive bond yields → GOLD is often under pressure. Conversely, as long as the market starts pricing in the direction of the Fed being less hawkish, the opportunity cost of holding gold decreases. This is exactly what is happening after the CPI. But this is only a short-term catalyst. But I don't think gold will go straight up. In early 2026, GOLD exceeded $5,500, then fell below $4,000 at the end of June. (World Gold Council) Such an amplitude says one thing: A bull market does not mean no correction. If the US economy is strong again + inflation heats up + FED hawkish + Treasury yield rises + USD strengthens, gold can be sold strongly. Conversely, if growth weakens, interest rate expectations fall, geopolitical tensions continue and Central Banks still buy... $4,400 may not be the top — it's just the GOLD zone trying to reclaim before a new bullish leg. The World Gold Council has assessed that strong enough catalysts can bring gold back to $4,500+ in H2/2026. (World Gold Council) If GOLD breakouts are confirmed by these areas with cash flows, the next narrative will not only lie in physical gold. It can spread to: $XAU → $XAUT → Tokenized Gold → RWA And as traditional haven assets begin to be brought on-chain, the line between TradFi and Crypto will become increasingly blurred. Don't just look at the price of gold. Look at where the big money is going. #GOLD #XAU #XAUT #BTC #RWA #FED #CPI #CryptoSudden uncertainties in Hormuz negotiations: BTC consolidates sideways, ETH weakly fluctuates, awaiting direction selection The US-Iran rivalry has escalated again, with the Hormuz navigation agreement, which was close to being finalized, officially put on pause. Iran has made it clear that as long as the US continues its military threats and fails to meet the conditions for unlocking assets and compensation, the Strait navigation agreement with Oman will be postponed indefinitely; Meanwhile, Trump's side has promised 'results within 48 hours' while countering compensation claims, while the US fleet remains deployed in the strait, and both sides have returned from 'close consensus' to a stalemate. As a result, international oil prices stopped falling and rebounded, and the interest rate cut trading logic that had been ignited by the unexpectedly falling in July CPI was once again overshadowed. Transmitting this to the crypto market, the two leading coins simultaneously entered narrow consolidation: • $BTC After the positive news materialized, it did not break through the 64,000 resistance level. It is currently oscillating between 63,300 and 63,600, with the 1-minute Bollinger Bands continuing to narrow. 63,000 serves as the short-term core support, and both bulls and bears are watching for geopolitical signals to materialize. • $ETH Weaker than BTC, currently quoted around $1880, 24-hour range narrowing to the 1873-1925 range, 1850-1870 as a short-term strong support zone, and 1920-1950 forming dense resistance. As a mainstream asset with stronger risk attributes, ETH's correlation with BTC remains at a high level of 0.9. During periods of rising geopolitical uncertainty, risk aversion will first suppress ETH's elasticity, and the decline is usually greater than BTC's; Conversely, if the situation eases, the rebound will be more explosive. #7月CPI平稳落地, expectations for a rate hike in September cooled The core logic remains unchanged: As long as there is no substantial upgrade in the short term, the market will continue to absorb profit-taking positions through volatility, with funds further concentrating on BTC's top positions, and ETH and altcoins will continue to face pressure; If an unexpected conflict occurs within the 48-hour window, a rebound in oil prices will reignite inflation concerns, causing overall market risk appetite to decline simultaneously; Conversely, if the protocol unexpectedly implements and the negative news is exhausted, the market will return to the main trend of rate-cutting trading, and ETH is likely to be the first to start a rebound and recovery. In terms of operations, it is recommended to wait and see for now. For BTC, focus on a breakout between 63,000 and 64,000 levels; for ETH, focus on the 1850 support and 1920 resistance. For high-leverage positions, prepare risk control in advance and wait for the range to break before following the trend.$OKB Major News $OKB AI Ecosystem Data, Traffic King OKB AI (X Layer AI Agent Ecosystem) On-Chain Daily Report [August 13], Underlying Logic of Price Increases 1. X Layer Total Network TVL: $117.9 Million, 7-Day Quarter-on-Week +9.1% ​ 2. Total unique addresses across the network: 4.2 million+, with total on-chain transactions exceeding 400 million ​ 3. Stablecoin Reserve: $2.08 billion, with Circle's native USDC continuously providing US dollar liquidity to the ecosystem ​ 4. OKB Fundamentals: Total permanently locked at 21 million tokens, no additional issuance; All on-chain interactions consume OKB as gas; the more interactions, the more deflation it sustains. 2. Data on the OKB AI Agent dedicated chain 1. AI Agent Contract Deployment Currently, the total number of AI Agent contracts deployed on X Layer chains is steadily increasing; At present, the focus is on developer testing bots and interactive agents, while large-scale, commercial, high-frequency calls are still in a gradual implementation phase. ​ 2. Gas consumption characteristics of the AI track AI bots are high-frequency, small-value interactions, with a single address generating hundreds to tens of thousands of on-chain operations daily; Similar to the 4.96 million interaction addresses you just checked, this is a typical AI script wallet. In the short term, the overall gas consumption of the AI sector is lower than that of the xStocks US stock token track; Large-scale AI returns have exploded, relying on a large number of ordinary users accessing the AI marketplace for sustained interaction. ​ 3. Ecosystem mechanism: OKB closed-loop consumption Developers deploying AI intelligent agents, AI human-machine dialogue settlement, and AI strategy execution all require payment of OKB Gas; In the future, after the AI service payment model is implemented, there will be a new demand for direct purchase of OKB. 3. Comparison of Capital Structures in the Three Major Tracks (Current) 🥇 Number one: xStocks US token RWA, accounting for 81% of on-chain DEX trading volume and currently the core source of OKB's consumption 🥈 Second place: DeFi exchanges and liquidity pools 🥉 Third place: OKB AI Agent Ecosystem (in early growth stage) 4. Key signals to monitor the AI ecosystem going forward 1. AI Marketplace officially opened on a large scale, adding a large number of ordinary user interaction addresses ​ 2. The proportion of daily on-chain transactions in the AI sector continues to rise, with gas consumption steadily rising ​ 3. Launched AI intelligent agent staking OKB feature, forming long-term chip lock-in. #Chip stocks lead the rally, Korean stocks rebounded over 22% in ten days, #7月CPI平稳落地 rate hike expectations in September cooled Yesterday it was $LITE, and today it’s $COHR. Since Maitong’s editor is off work, I’ll get the update out first. 😄 $COHR’s earnings were just as impressive: 1. Strong earnings beat: Revenue came in at 20.46 vs. 19.81 expected, with solid growth in data center and communications. Non-GAAP gross margin jumped to 40.2%, while EPS reached 2.41 vs. 2.19 expected. 2. Guidance also beat: Q1 revenue is expected at 22–24 vs. 21.5 forecast, with EPS guidance of 1.85–2.05 vs. 1.79 expected. In simple terms, FY2026 Q4 is already strong, but FY2027 could be even more impressive as growth accelerates. Both $LITE and $COHR are delivering strong signals from both fundamentals and technicals. So the bigger question is: Could optical stocks become the next major hardware theme after memory/storage? I think that possibility is becoming increasingly obvious. Personally, I also feel that the US market is still much more enthusiastic about hardware-related speculation than software. Software currently feels like it’s stuck waiting for the next narrative and new buyers. Once that demand fades, the sector could lose its appeal quickly. Just my personal view. #KoreaChipsLeadRebound #CPIEasesHikeBets #AnthropicIPOValuation At a SpaceX all-hands meeting, Musk put out a big satellite: next month, AI revenue will surpass all other businesses combined, and by the end of next year, it will reach 10 gigawatts of computing power, earning $300 to $500 billion annually. In five years, AI will account for 99% of the company's value—this is the plan to turn Rocket Company into an AI giant. The current approach is called "ground training, space reasoning": training stays on Earth, reasoning moves into space. Starship is responsible for delivering computing hardware, Starlink handles the communication base, forming a comprehensive approach. On paper, the AI business is indeed strong—AI revenue in Q2 was $2.6 billion, up 213% quarter-on-quarter. But the company posted a net loss of $4.8 billion in the first half, and just the computing power investment in AI has eaten up all its cash flow. For $BTC: Musk's dream this time is bigger than the "Tesla robot"—if the AI + space story succeeds, $SPCX will become the top liquidity pump in the US stock market, drawing in tech capital. The bigger problem is that if he really relies on AI to push valuations to "astronomical figures," the entire market's risk appetite will be redefined, and BTC's "tech asset" attributes will be shaken. Let's take a look first. #马斯克称AI将占SpaceX价值99% $ETH — Why Can’t Ethereum Hold Above $1,900? On August 13, $ETH stayed mostly between $1,875–$1,895, barely moving over 24 hours. After briefly touching $1,934, it quickly pulled back and failed to reclaim $1,900 for the third time. Although cooler CPI data and reduced September rate-hike expectations should have supported ETH, the bullish momentum simply isn’t strong enough. The US spot ETH ETF attracted only around $7.4M in net inflows on August 12, far below previous levels. Without stronger ETF demand, it’s difficult to absorb the heavy supply around $1,900+. Technically, ETH remains stuck in a consolidation range, with Bollinger Bands around $1,839–$1,946 and weak ADX showing limited momentum. Key levels: Above $1,900 with strong volume: Potential move toward $1,920–$1,930 Below $1,870–$1,880: Next targets around $1,850, then $1,820 Between these levels: Likely just sideways chop and short-term position shakeouts. In simple terms, ETH isn’t necessarily showing strong selling pressure—it’s simply lacking enough buying power to break higher. Not financial advice. $ETH #KoreaChipsLeadRebound #CPIEasesHikeBets #AIInfraEarningsWatch Bridge exploits keep proving that the wrapped token is only as safe as the accounting behind it. 🌉🚨 Tx says an attacker used fake XRPL deposits to withdraw real XRP from its reserves, while other bridged assets remain fully backed. The key failure appears to be deposit verification, not XRP itself. Before trust returns, I’d want proof of reserves, the exact amount drained and an explanation of how false deposits passed validation. Other assets are backed is useful but independently verifiable backing is better. $XRP $NOT $CATI #CPIEasesHikeBets #AIInfraEarningsWatch #SpaceX99%ValueFromAI $ONDO Grvt could make Ondo far less visible to the end user. The headline will probably be: “$100M allocated to tokenized Treasuries.” But the more interesting part is what happens behind the scenes. Grvt is reportedly holding USDY on its balance sheet and integrating the yield directly into a base rate. That means users can potentially earn a Treasury-like return without ever needing to interact directly with the token or Ondo. That’s a bigger shift for RWAs. The product is no longer “buy this token.” It becomes: “the yield is already built in.” The numbers are worth watching: • $100M targeted over 12 months • Around 4.6% of existing USDY supply • Roughly 3.5% APY, translating to about $3.5M annually at full deployment One platform potentially representing nearly 5% of a $2.1B market shows just how concentrated onchain fixed income remains. But there’s another side to this. Abstraction makes things easier for users, but it can also hide the underlying complexity. The user may simply see one clean rate while the backend involves multiple layers—Grvt, Ondo, banks, ETFs, and other counterparties. The real winners in the RWA cycle may not be the platforms offering the highest yield. They’ll be the ones that make the infrastructure almost invisible without making the underlying risks invisible. Tokenized Treasuries are moving beyond being a standalone product. They’re becoming a built-in feature. And that shift could be much bigger than most people realize. Who else sees it this way? Video credit: @new_era_finance #CPIEasesHikeBets #AIInfraEarningsWatch #SpaceX99%ValueFromAI A brother asked in the comments whether Core’s GitHub being quiet for a long time is a warning sign, so I looked into it. My view: fewer code updates deserve attention, but they don’t automatically mean Core is failing or abandoning development. A public-chain codebase doesn’t need daily commits. Once the core infrastructure becomes stable, development may focus on security patches, maintenance, upgrades, or other repositories. The real concern is when low code activity comes together with fewer developers, delayed mainnet upgrades, weak security development, shrinking TVL, fewer ecosystem projects, and declining active users. That combination would be a much stronger warning. So instead of only asking, “Has Core’s GitHub been updated recently?”, look at the bigger picture: Are technical upgrades still happening? Are new versions being released? Are core developers active? Is the mainnet progressing normally? Are ecosystem activity, TVL, and users growing? I checked Core DAO’s official GitHub, and it’s too early to say that development has completely stopped. Other official repositories have also seen activity this year. So don’t panic based on one metric. As a $CORE holder, I’ll definitely keep watching the development closely. Code activity alone doesn’t determine whether a blockchain is alive—but if development, ecosystem growth, TVL, and users all start declining together, that’s when the real alarm bells should ring. #KoreaChipsLeadRebound #CPIEasesHikeBets #HarmonyMintRollback 📊 $ETH Contract Overload Express (August 13) According to liquidation data, ETH shows a pattern of repeatedly switching between multiple cycles and directions, with extremely intense bullish and bearish tug-of-war: · Short Cycle (1H): Total liquidation $33,600, long position $29,600, short $4,004.59, bulls crushing bears by 7.4 times, bullish selling dominates but the volume is small. · Medium and short cycle (4H): Total liquidation $2.44 million, short $2.1087 million, long $331,400, bears crushing bulls 6.36 times, sharply reversing direction, short squeeze market explosive at 4-hour level. Short-term cycle double kill characteristics for both long and short are pronounced. · Medium cycle (12H): Total liquidations $7.2962 million, long positions $4.3986 million, short positions $2.8976 million, bulls crushing shorts by 1.52 times, direction reversed again, long selling trend returned, liquidated positions about double compared to 4 hours. · 24-hour timeframe: Total liquidations $36.1671 million, short positions $19.6982 million, longs $16.4689 million, bears overtaking bulls **1.20x**, direction reversed again, short squeeze dominated the 24-hour chart, cumulative liquidations broke $36.1671 million, short positions accounted for nearly 54.5%. ⚠️ Risk warning: ETH repeatedly switches in multiple cycles (1H long sell→ 4H short squeeze→ 12H long sell→24H short squeeze), with a typical pattern of quadruple kills for long and short positions; Over $36 million cumulative liquidations in 24 hours, market volatility is intense but direction is highly unclear. Leverage is recommended to be compressed to within 3x; do not chase gains or sell, strictly control positions, and wait for clear direction. 🔥 Market Barometer | August 13 Today's three hot topics point to the same theme: AI narratives are undergoing performance validation from "burning money" to "making money," and the macro environment is simultaneously providing a window for this validation. 📊 July CPI Landing Steadily: Expectations for a September Rate Hike Fad US July CPI was 3.4% year-on-year and 0.1% month-on-month; Core CPI was 2.5% year-on-year and 0.2% month-on-month, all fully in line with expectations. The main drag was the drop in fuel prices, with gasoline prices falling 2.9% month-on-month; The month-on-month increase in food items narrowed to 0.1%, while lettuce and tomato prices, which had previously been affected by supply shocks, plummeted. After the data was released, the probability of a rate hike in September plummeted from 48% two days earlier to 36%. Nick Timiraos, known as the "new Fed News Agency," pointed out that the report "has somewhat eased pressure for the Fed to raise rates next month." The S&P 500 index closed up 0.3%, near its all-time high. 🏗️ AI infrastructure earnings relay: cloud revenue accelerates across the board During the Q2 earnings season, the three major cloud providers delivered strong results. Google Cloud's revenue was $24.8 billion, an 82% year-on-year increase, with its operating profit margin jumping from 20.7% to 35.6%; Microsoft Azure grew 43% year-on-year; Amazon AWS revenue was $42.2 billion, up 37% year-on-year. All three major cloud providers achieved more than double their unfulfilled orders. AI investment is forming a positive cycle of "capital expenditure→ revenue→ profit → further increase." The rising AI cloud infrastructure star also exploded—Nebius's core AI cloud business sales surged 514% year-on-year, with its stock price soaring 34% in a single day; CoreWeave disclosed $104 billion in orders on hand, with its stock price surging over 19%. 🚀 Elon Musk: AI will account for 99% of SpaceX's value At the all-hands meeting, Musk boldly declared: AI revenue will surpass all other SpaceX businesses as early as September; Within five years, AI will account for 99% of the company's value; SpaceX aims to build 10 gigawatts of AI computing power by the end of next year, corresponding to annual revenue of $300 billion to $500 billion. SpaceX's current computing power is 1.4 gigawatts. Boosted by this, SpaceX's stock price rose over 6%, rebounding more than 35% from previous lows. 💎 Summary CPI is moderately implemented, with the probability of a rate hike in September dropping to 36%, temporarily easing macro pressures; The three major cloud providers have demonstrated that AI investment is paying off with operating profit margins exceeding 35%; Elon Musk, on the other hand, has declared, "AI accounts for 99% of SpaceX's value," pushing the imagination of AI narratives to new heights. As the macro window opens, the industry cycles are established, and the narrative ceiling is redefined—the AI track is moving from "storytelling" to a stage of "delivering the answer sheet." #7月CPI平稳落地, expectations for a rate hike in September have cooled down #财报观察员: AI infrastructure earnings report debuts one after another #马斯克称AI将占SpaceX价值99% $SOL Rob has been bearish on Solana for months, but a recent post from @lukemartin made him rethink part of his thesis. His earlier view was that Solana was still too closely associated with Pump.fun and meme-coin activity, making it difficult to attract serious institutional capital or gain meaningful market share in more established financial markets. But there’s another possibility. If tokenized stocks and other real-world assets become widely used onchain, the same distribution and speculation mechanics that made Pump.fun successful could potentially work for a much broader audience—especially younger investors who aren’t interested in traditional meme coins. That doesn’t necessarily mean being bullish on Solana as an L1. It means there may still be a legitimate path for the ecosystem to succeed through tokenized assets and new forms of onchain trading. Andy’s counterpoint was even more interesting: if that’s the thesis, why buy $SOL when $PUMP could potentially capture more value after accounting for Solana’s inflation and token emissions? The debate now isn’t simply whether Solana wins. It’s which part of the Solana ecosystem actually captures the value. #CPIEasesHikeBets #AIInfraEarningsWatch #SpaceX99%ValueFromAI $APR Why do so many people always say there are many short sellers? If there really were so many short sellers, the funding rate wouldn't be positive. In the past 24 hours, short positions have clearly dominated. So, the next rally will depend on who moves faster. Now it's basically the final round. In the bullish battle, the bears have already broken up wherever they can. The new investors have leveraged more than three times and have ample margin. If they get liquidated, the main players will probably have a headache watching this. Pulling them up and recklessly pursuing them is unrealistic. Yesterday, it rose 220%, but the short sellers only saw a little over six million in losses. However, the transaction volume exceeded $800 million. It's easy to imagine how difficult it is to settle accounts upward. Therefore, a rally could happen at any time, but the likelihood of a short strike is reduced, with more distribution aimed at high levels. Don't talk about 'monster stocks' or not. When leveraged is pulled up, the higher the price, the greater the risk for the main players. After last night's rally, the main players' costs should be quite high now. They probably struggle to sell shares, wanting to sell at high levels, but there are too many chips, making it hard to sell. Also, there was no shakeout at launch, with a bunch of positions with gains of ten or twenty times the float. If you push higher, you can sell, making it harder for the main players to sell. At the same time, there is concern about whether short-selling institutions might be targeting them. Therefore, trading time for space will eventually evolve into distribution regardless of cost. Resistance above: around 0.53. Volume breakout is expected to reach 0.65. Resistance below: 0.43. If it breaks out on increased volume, it won't be possible to see the bottom. #7月CPI平稳落地, expectations for a rate hike in September cool down#财报观察员: AI infrastructure earnings report debuts in succession, #马斯克称AI将占SpaceX价值99% $FIL bulls look extremely weak right now, while bears seem to have far more selling power. A single dump of around 200,000 can push the price down by roughly 30 points, so it’s hard to imagine what would happen if sellers unloaded 2 million—the price could potentially fall toward 0.64. Because of this imbalance, $FIL looks more suitable for short setups than long positions. Most long buyers appear to be retail traders without enough capital to absorb heavy selling, while bears seem capable of unloading tens or even hundreds of thousands in one move. For now, it may be better to stay away from this coin unless the market structure changes. The bulls simply don’t seem to have strong enough liquidity or market support to compete with the selling pressure. #CPIEasesHikeBets #AIInfraEarningsWatch #SpaceX99%ValueFromAI Knockoffs collectively stalled, $OKB bucked the trend to break through $100, rising nearly 8% in 24 hours! $BTC remains near $63,000, $ETH holds steady at $1,900, and $SOL hovers around $76. Mainstream coins have temporarily stabilized, but sentiment in the altcoin market has clearly weakened. The NFT sector fell 5.19% in 24 hours, with $BEAT dropping another 13.31%. The Meme sector fell 3.69%, with $PEPE and $TRUMP falling by 5.61% and 6.58%, respectively. The recession of previously high-volatility sectors indicates that funds are indeed actively narrowing their holdings. A few coins are still operating independently. The most eye-catching is $OKB, which fell 0.52% overall in the CeFi sector, while $OKB rose nearly 8%. Recently, $OKB has shown strong performance, possibly because the market is betting on the funds and applications that X Layer has recently replenished: - Circle connects native USDC and the cross-chain transfer protocol CCTP to X Layer - Pendle completed native X Layer deployment and launched the USDG yield market - The scale of stablecoins on X Layer is about $2 billion, and DeFi TVL has surpassed $100 million, with nearly tenfold growth in half a year If more lending, trading, and yield protocols join next, funds will truly flow on X Layer The weak narrative continues to bleed, with a few strong coins draining liquidity. Choosing the wrong direction can be even worse than missing out.Don’t play the Monday-morning quarterback—tonight’s PPI is also likely to come in close to expectations. Yesterday, July CPI showed 3.4% YoY growth and just 0.1% MoM, confirming that inflation is gradually cooling. For tonight, PPI is expected at 4.9%, down from the previous 5.5%. If the figure matches forecasts, it would simply reinforce the cooling-inflation narrative from CPI rather than deliver a fresh surprise. After the CPI release, $ETH jumped to around $1,924.97, only to quickly drop more than $70 toward $1,870. That’s a classic “buy the rumor, sell the news” reaction. So even if PPI meets expectations, I don’t expect it to create another strong, sustained rally. The more likely scenario is another quick pump followed by a pullback. #KoreaChipsLeadRebound #CPIEasesHikeBets #HarmonyMintRollback From the "retail investor graveyard" to the tech bull market, the Korean stock market took only 10 days—Old Mo tells you why chip stocks pulled the KOSPI back Guys, this V-shaped reversal in the Korean stock market made Lao Mo say it was ruthless. On August 13, South Korea's KOSPI index surged as much as 4.8% intraday, rebounding about 22% from the July 30 low, officially entering a technical bull market. Ten days ago, panic selling was underway; ten days later, it directly entered the bull market. Who pulled it up? Chip stocks. Samsung Electronics rose over 5%, and SK Hynix gained over 7%. Both companies contributed the most to the KOSPI index. Storage concept stocks rebounded collectively—SK Hynix rose over 9%, Seagate Technology rose over 7%, SanDisk rose over 5%, and Micron Technology gained over 4%. So far this year, the KOSPI index has risen more than 60%. After a sharp 22% drop in July and marking the worst monthly performance since the global financial crisis, it was fully recovered within 10 days. Why can it come back? Lao Mo will break it down for you with four reasons. First, AI storytelling is back. Earnings reports from CoreWeave and Supermicro confirm strong demand for AI infrastructure, directly igniting market enthusiasm for tech hardware stocks. Global tech giants continue to show large-scale AI spending in their latest earnings, rebuilding market confidence in memory chip demand. Mark Newton, Head of Technology Strategy at Fundstrat, said that memory chip stocks have begun to outperform the broader tech sector for the first time since June—"This is significant for South Korea, as Samsung and SK Hynix have a significant impact on the Korean stock market." Second, those who were liquidated by leverage have disappeared. The core reason for the July crash was the concentrated liquidation of leveraged chip stocks, which led to trading halts and wiped out billions of dollars in wealth from South Korean retail investors. The South Korean government subsequently tightened measures related to leveraged ETFs for individual stocks, and investors reduced margin debt. Arkevium Capital's Chief Investment Officer put it bluntly: "Once leverage is removed, the same market can experience a sharp rebound." Investors forced to sell disappear. Short sellers take profits. Traders reduce downside hedging. ” Third, CPI provided a boost. US July CPI year-on-year slowed to 3.4%, and core CPI slowed to 2.5%, all in line with market expectations. After the data release, the probability of a rate hike in September dropped from nearly 50% to around 38%. Concerns over Fed rate hikes eased, providing support for US-listed chip stocks, which in turn passed on to the Korean stock market. Fourth, Temasek and shareholder return plans are being ignited. Reports say Singapore's state-owned investment company Temasek plans to invest in South Korea's chip giant, causing Samsung and SK Hynix's stock prices to surge more than 8%. At the same time, the market expects Samsung and SK Hynix to announce new shareholder return plans as early as the end of August, with a combined return size possibly exceeding 200 trillion Korean won (about 952 billion yuan). Can the rebound last? Lao Mo mentioned a few signals. Life Asset Management CEO Kang Dae-won reminded: "If the AI investment narrative and U.S. interest rate trends cannot achieve a certain degree of stability, it will be difficult for the Korean stock market to sustain sustained gains." ” Fundstrat believes there may be further room for a rebound—the iShares MSCI Korea ETF has broken through key technical levels, confirming a reversal pattern. Macquarie analysts maintain the KOSPI index target of 8,000 points. But so far this year, foreign capital remains a net seller, with over $100 billion expected to withdraw from Korean stocks by 2026. Foreign capital has yet to return; this rebound mainly relies on domestic capital and short covering. Back to the big cake ether. This volatility in Korean stocks isn't directly related to your Bitcoin orders, but there are two indirect signals worth watching. First, the AI narrative is being regained market recognition, and risk appetite in the tech sector is rebounding—as a high-beta asset, Bitcoin will benefit emotionally. Second, if the Kimchi Premium turns positive as the Korean stock market rebounds, the selling pressure from Koreans may ease. Over the past month, the reverse Kimchi Premium has been suppressing Cake, and this variable is worth watching. BTC's latest price is around 63,700-64,200, fluctuating within the 63,500-64,500 range within 24 hours. ETH is quoted at around 1890-1910. Lao Mo concludes: From a 22% plunge to a 22% rise in 10 days, the Korean stock market has proven one thing with action—the AI narrative isn't dead, it's just that the leverage stampede in July pushed prices where they shouldn't be. But the rebound ≠ reversal, foreign capital hasn't returned, and the sustainability of AI spending remains controversial. KOSPI is still about 24% away from its June high—there's still a long way to go. Did you feel the Korean stock market rebound this time? Will the big cake follow suit and drink the soup? Let's talk in the comments. #芯片股领涨, Korean stocks rebounded over 22% $BTC $ETH $OKB in ten days #特朗普因TruthSocial付费数据流遭起诉 Sue the President? My first reaction was: if it's just about spending money on a better API, I think it's perfectly normal; but if you can pay to buy "president's message a few milliseconds faster," then the nature is completely different. At the heart of this controversy is Truth Social's launch of the Truth API. According to Reuters, the service charges up to $100,000 per month, allowing buyers to access posts from high-influence accounts like Trump faster than regular users; The Intercept and the Freedom of the Press Foundation have filed federal lawsuits, arguing that when these content involves tariffs, sanctions, and other policy information that could directly impact the market, paying users may gain a trading advantage. We all know that when dealing with someone like Trump, who can influence $BTC, US stocks, or even crude oil with a single word, a few seconds can be enough to push the price upside by bit. Ordinary people see the push and then open trading software; quantitative funds might have already completed the first round of trading through the API. So I don't oppose paid data services; Bloomberg, exchange quotes, and institutional terminals naturally have speed gaps. But if the information itself is a public release by the president and may affect financial market policies, then "whoever pays more, knows first" easily crosses the boundaries of market fairness. This has made me less and less chasing the first piece of news now. Retail investors find it hard to beat institutions in speed, so don't compete with machines for speed. I'd rather wait until the first wave of emotional release to judge whether this news has changed the fundamentals or just created a few minutes of volatility. In a market driven by increasingly strong news, being a few seconds behind doesn't necessarily mean losing out; chasing the wrong direction is the real lossThe Korean stock market has been rebounding fiercely this time. In ten days, it hit 22 points, moving directly from a technical bear market into a technical bull market. Samsung rose more than 5%, SK Hynix rose more than 7%, and the sharp rally even triggered a pause in programmatic buy orders. The more badly the previous drop, the stronger the rebound now. The logic is actually quite clear: AI capital spending continues, storage and optical communications sectors are recovering, and foreign capital is flowing back. Temasek is reportedly considering direct investments in Samsung and SK Hynix. Although the timing and scale are yet to be determined, the news alone is enough to fuel sentiment. What does this have to do with the crypto world? South Korea is one of the world's most active crypto markets. Aggressive stock market rebounds will draw away some capital, and the Korean premium has recently narrowed, which is partly related to this rebound. In the short term, Korean funds in the crypto market will come under pressure. But chip stocks leading the rally show that the AI narrative is still in effect. As long as this line doesn't collapse, the overall risk appetite in the tech sector will not be too bad. As a high-beta asset, crypto will indirectly benefit sentimentally. Besides, the Korean stock market's performance itself is a signal. A 17% drop in ten days, then a 22% rise in another ten days—this level of volatility can't be explained by fundamentals—it's the result of leveraged clearing and covering. South Korean retail investors are used to high volatility, and after this stock market rebound, funds are very likely to flow back into the crypto market. Let me share my thoughts. This rebound is essentially a recovery after an over-dip, not a sudden improvement in fundamentals. The storage boom still exists, but previous valuations were hit by heavy drops and pits; now only part of the hole has been filled. Samsung and SK Hynix alone hold a large weight in the index; as soon as funds come in, the index soars, rising quickly and falling quickly. For the crypto market, short-term capital flow pressure remains unchanged, but in the medium to long term, the logic behind South Korea's capital allocation remains unchanged. After the stock market rises, what should return will return. $BTC $ETH #芯片股领涨, Korean stocks rebound over 22% in ten days The early morning square is even more exciting than the daytime candlesticks. $APR jumped from 0.21 to 0.63, then back to 0.48, killing all bulls and bears in a single day. The 10x short guy entered at 0.397, with a floating loss of -435%, still holding on; Then trader Maomao went all-in on 12wu, reasoning "this candlestick looks too much like a pullback to attract bears." The same coin, two faiths—one side will have to pay tuition. If you've seen the Yaobi game a lot, you'll understand: the $BEAT comes $BICO, the $BICO cools down and $APR comes back, with a new protagonist every day. The same group is pulling the market, while the new ones taking over are coming in wave after wave of newcomers. Maomao dares to bet on 12wu because the SNDK order still holds onto a 100,000 U floating profit as a base—they can afford to lose, but if you follow the crowd and lose, no one will cover the loss. The market was quiet: $BTC 63,800 had been grinding for three days, $ETH 1890 was hovering, $SOL hovered around 76. Funds haven't left; they're all rotating within the 'demon coins.' Once this wave of sentiment is digested, the main theme will return. $BTC $ETH $SOLI don’t think AI agents create automatic demand for every AI coin. The payment loop has to touch the chain. 🤖⛓️ Grayscale expects agents to drive demand for Ethereum and Solana in finance, Worldcoin for identity and Bittensor for decentralized AI governance. My filter is simple: does the agent actually pay fees, hold collateral or verify identity on-chain? If yes, usage can create structural demand. If the token only sits beside an AI narrative, the connection is mostly marketing. #CPIEasesHikeBets #AIInfraEarningsWatch #SpaceX99%ValueFromAI $BTC $ETH $APR CoreWeave’s revenue growth caught my eye, but the commitments matter even more. ⚡ The company reported $2.58B in Q2 revenue, up 112% year over year, and signed over $25B in new customer commitments in early Q3. That suggests AI compute demand is being booked years ahead, not bought casually each month. The next question is execution: commitments only become valuable if CoreWeave can finance and deliver the required capacity without infrastructure costs outrunning revenue. #CPIEasesHikeBets #AIInfraEarningsWatch #SpaceX99%ValueFromAI $BTC $ETH $OKB $OKB This rally actually has logical support First and foremost, the core issue is the restructuring of scarcity. Previously, OKX burned about 65.26 million OKB at once, permanently fixing the total supply at 21 million. After a sharp contraction in supply, the valuation logic for OKB has changed  The second catalyst is X Layer. Now, OKB has become the core gas asset of X Layer. As OKX further integrates transactions, wallets, payments, and the on-chain ecosystem, its practical use cases are continuously expanding  More notably, ICE, the parent company of the NYSE, invested in OKX this year, pushing OKX's valuation to around $25 billion. The two parties subsequently established a joint venture to continue advancing tokenized assets and digital financial infrastructure  So now, OKB is no longer just an exchange platform token, but more like the core asset of the entire OKX on-chain ecosystem. With reduced supply and increased applications, combined with the entry of traditional financial giants, market funds naturally find it easier to re-price it. Recently, I've been trading US stocks and oil. Although I'm also working on dual-currency BTC, I mostly focus on ETFs and volatility, and have indeed overlooked BTC's on-chain data. I checked during the day today and can only describe it as unbearable to watch. Indeed, on-chain BTC data is very poor, especially some of the data I focus on personally, which is very poor. For example, exchange stock data has always been my focus. After bottoming out on May 5, 2026, the exchange stock increased by more than 130,000 BTC in just three months. This data basically shows that although Bitcoin has been hovering around $60,000, some users are still transferring BTC to exchanges to prepare for selling. From the data, Binance has the largest transfer volume, with over half, followed by Coinbase and OKX, each with over 20,000 Bitcoins transferred. Although this amount of inflow does not mean a large amount of chips are waiting to be dumped immediately, the BTC that has been flowing in since the $80,000 drop may be sold off during the rise. In other words, it's highly likely that the current $60,000 is not the target price for these holders, but the closer BTC gets to $80,000, the greater the potential sell-off it will face. $BTC Everyone, South Korea's KOSPI index has rebounded violently 23% from its July 30 low, directly entering a technical bull market. Samsung Electronics and SK Hynix, two heavyweight stocks, rose over 5% and 7% respectively today. This round of rebound is trading a triple resonance. The first clear card: Temasek is about to "bottom-fish." Singapore's sovereign wealth fund Temasek plans to invest directly in the Korean stock market for the first time, targeting Samsung Electronics and SK Hynix. This is Temasek's first entry into the Korean stock market, and the signal is more important than the money itself. As soon as the news broke, both Samsung and Hynix's stock prices surged over 8% that day. The second clear card: the largest shareholder return in history. SK Hynix is preparing a shareholder return plan totaling about 100 trillion won (about 71 billion USD), including 40 trillion won in buybacks and 60 trillion won in special dividends. Compared to last year's roughly 14.3 trillion won, this is a nearly sevenfold increase. CLSA Securities believes shareholder returns in 2026 could exceed 100 trillion won, with a target price of 3.7 million won. Samsung may follow, with the two companies together possibly exceeding 200 trillion won. The third clear card: the supply-demand gap "has no end." Micron executives publicly stated at the KeyBanc forum that storage supply and demand in 2027 will be tighter than in 2026. The exact words are, "The primary limiting factor is DRAM itself, not power or factories." JPMorgan has already raised its forecast for the global storage market size from 2026 to 2028 by 4% to 8%, projecting it to grow from $969 billion in 2026 to $1.44 trillion in 2027 and $1.82 trillion in 2028. Three "obvious" positive factors were laid out simultaneously: strong fundamentals, shareholders willing to share, and foreign giants endorsing them. This triple resonance propelled KOSPI into a technical bull market. But to be honest, the cyclical nature of memory chips hasn't disappeared; they're just temporarily masked by AI demand. Right now, the forward P/E ratio is indeed cheap, but cheapness is never a reason to rise—it's the poor expectations that matter. This round of rebound trading is 'fundamentals remain firm + shareholders are willing to share money + endorsement from foreign giants,' and the logic is quite smooth. But the storage industry's 'cycle curse' still hangs around; once the expansion cycle starts, the supply-demand gap will be filled sooner or later. But for now, that time hasn't come yet. Have you kept up with the pace this time? Share your moves in the comments. Wishing everyone smooth trading $BTC $SKHYNIX $SNDK Title: Harmony Faces a "Money Printer" Attack—Is Rollback the Antidote or Poison? Event Overview: The Harmony network experienced an anomaly today, with hackers generating over 3 trillion ONE (several times the normal supply) through a minting vulnerability, involving six blocks. The authorities have confirmed and urgently activated a rollback plan; the vulnerability has been fixed, and the list of attacker wallets will soon be released. Key highlights: 1. What does rollback mean? This is equivalent to "changing history," restoring on-chain state to the pre-attack state. This requires validators and exchanges to jointly execute. While it can erase hacker profits, it also shakes the foundation of "immutability," potentially causing community consensus to split later. 2. Market Impact: · Short term: If the rollback succeeds, selling pressure disappears and prices may rebound, but liquidity may be limited due to exchange suspensions. · Long-term: Public chain security and governance mechanisms are being questioned. If a fork occurs, how to identify new and old tokens will be a huge risk. 3. Operational Recommendations: If you are considering short-term longing, please pay close attention: · During the rollback, the price is very prone to "pin insertion" up and down, so it is recommended to set a hard stop-loss position; · Pay attention to official announcements and the timing of exchange resumption of deposits and withdrawals; · Do not heavily bet on a rebound, as such events carry extremely high uncertainty. Summary: Harmony's choice of a "hard rollback" this time is a last resort. It can save lives in the short term, but in the long run, it depends on whether the community buys in. Small positions are fine, but don't take the rollback as a positive sign. Personal plan: Small positions with orders to go long, stop loss set at 5% below the pre-rollback low, exit immediately after the price drops, no pattern. · Remember to pay attention to when exchanges resume deposits and withdrawals, as this is often the window of maximum price volatility.如果说英伟达卖的是AI的“大脑”,那么美光科技($MU)卖的,就是AI运行过程中越来越不可缺少的“记忆”。 过去市场看美光,通常把它归类为典型的周期股:内存价格上涨,公司赚钱;供给增加、价格下跌,利润又被打回原形。 但2026年的美光,正在发生一个非常关键的变化: AI正在把传统内存周期,改造成一场更长期的算力基础设施扩张。 🚀 美光最核心的逻辑:AI正在疯狂“吃内存” 现在AI模型越来越大,数据中心部署的GPU越来越多,对高带宽内存(HBM)、服务器DRAM以及高性能存储的需求同步增加。 美光最新财报已经把这个趋势表现得非常明显。 2026财年第三季度,美光营收达到414.6亿美元,相比上一季度的238.6亿美元大幅增长;GAAP净利润达到282.4亿美元。公司同时给出的第四季度营收指引约为500亿美元,毛利率预计约86%。(Micron Technology) 这已经不是传统意义上的“内存周期反弹”。 AI正在把内存从普通零部件,变成算力基础设施的核心资产。 🧠 HBM,可能才是美光真正的王牌 为什么市场如此关注美光? 核心就是两个字: HBM。 HBM,也就是高带宽内存,是AWoke up to a breakthrough over 100U, not just pure sentiment, mainly because: 1. Technical breakout after weeks of consolidation (82-87 range), with volume pushing through the 90-100 psychological barrier 2. Circle native USDC + CCTP officially launched on X Layer, increasing demand for OKB as a Gas token 3. OKX ecosystem moves (European rewards, product expansion) + altcoin rotation capital inflow#CPIEasesHikeBets #AIInfraEarningsWatch #SpaceX99%ValueFromAI $XRP and 1001 questions: WHY? 1. Cheap gas fee + very little burn: Although the volume is high, the amount of burning is very low, not enough to create scarcity >> The price does not increase 2. Revolving design, the same amount of $XRP can serve many transactions in a day >> The faster and cheaper it is, the more difficult it is to push the price up (Funds, organizations, individuals do not need to keep too much $XRP to pay gas fees) 3. No staking, weak yield, very low reserve (1 XRP/1 account) 4. Supply is always higher than demand: Unlock new tokens every month ... ... #RippleOnce the hype from an Ansem post stops attracting fresh buyers, $Hx tends to drift lower. He started promoting WIF around a $100K market cap, before it eventually pushed above $3B, while his audience had grown to roughly 600K followers by late 2024. The key question now is whether organic buyers can keep showing up even when there’s no new post or promotion. If demand only appears when another call goes viral, the rerating may struggle to hold. But if buyers continue accumulating during the quiet periods, the move has a much stronger foundation. #CPIEasesHikeBets #AIInfraEarningsWatch #SpaceX99%ValueFromAI "Bitcoin ETFs saw a net inflow of $850 million last week, so why is BTC still hovering around 64,000?" ETFs are buying, but whales and miners are selling. Both sides are fighting each other, and who wins is still unknown. Let's first look at ETFs. Last week, US spot Bitcoin ETFs saw net inflows for five consecutive trading days, totaling about $853 million. BlackRock's IBIT alone attracted $694 million, accounting for over 80% of the total. BlackRock himself says: Bitcoin ETF investors tend to hold long-term. Even though many bought near $100,000 or even $110,000 and have already posted clear losses, there is still no panic retreat. The toughest institutions on Wall Street are accumulating shares bit by bit at the $60,000 mark. But looking at the other side. On-chain data is sending alerts. Lookonchain monitored that an anonymous whale sold a total of 7,513 bitcoins over the past three weeks, with a total value of approximately $486.9 million. Another whale, suspected to be a miner, deposited a total of 6,494 BTC to Binance over the past 20 days, worth $421 million, with an average price of $64,798. The most heartbreaking part is that one of the addresses' BTC was received from FalconX a year ago at an average price of $116,110. Held for a year, lost 44%, then sold at a loss. You think only retail investors are losing money? Big players are bleeding just as much. So the current situation is: ETFs are buying, whales are selling. Miners are turning, big players are running. Buying comes from long-term allocation funds on Wall Street, while selling comes from on-chain stock selling pressure. The two forces are facing off near $64,000. BTC is trading sideways at this level, with volatility dropping to its lowest point of the year. No one has won. But something interesting happened— ETH and BTC have started to go their separate ways. Data from August 10 shows that Bitcoin ETFs saw a single-day net inflow of 1,731 BTC (about $112 million), while Ethereum ETFs saw a single-day net inflow of 29,900 ETH (about $56.78 million). Looking at the past 7 days, Ethereum ETFs have accumulated a net inflow of about 118,500 ETH, valued at approximately $225 million. Although the absolute amount is not as high as BTC, ETH is more efficient at accumulating shares in terms of scale. Moreover—BTC has dropped nearly half, ETH has fallen from 4,800 to 1,900, a 60% decline. Which is cheaper? Speak for itself. There is another variable. On August 12, the US July CPI data will be released. Economists expect overall inflation to slow to 3.4%. If CPI continues to cool down, rate cut expectations heat up, and risk appetite reopens—ETH is usually more resilient than BTC. If CPI exceeds expectations and rate hike expectations heat up—then no one can escape it. The current location is the calm before the storm.U.S. stock funds are withdrawing from most industries, yet continue to pour money into tech stocks Recently, U.S. retail investors have started to reduce purchases of individual stocks, even net selling, but their net purchases of ETFs remain relatively stable. Retail investors have not truly left U.S. stocks; they have only started to reduce risk, shifting from pursuing one stock to betting on the continued rise of U.S. stocks as a whole. Recent BofA data further supports this view: over the past week, BofA clients net bought about $3.8 billion in U.S. tech stocks, the second-largest single-week purchase in history. However, at the same time, 7 out of 11 industries saw net selling, while all U.S. stocks combined for a net sale of $2.4 billion. The industrial sector saw a net sale of about $1.9 billion, communication services a net sale of $1.8 billion, financials a net sale of $1.4 billion, and healthcare a net sale of $1.3 billion. In other words, there has not been a widespread capital inflow back into the U.S. stock market; instead, buying interest is increasingly concentrated in tech stocks. Retail investors are reducing their positions in highly volatile individual stocks, shifting more money into ETFs, while others actively concentrate their holdings in tech companies. The result is actually very close: funds are increasingly flowing into the leading companies with the highest index weights, the best liquidity, and the strongest market consensus. So now, a fairly obvious situation may emerge in US stocks: the index itself remains strong, but many stocks below the index have not received the same capital support. $QQQ Why did the prototype of the big short seller in 2008 choose to keep increasing his position, shorting Nvidia and related semiconductors? He believes that NVIDIA is not just a company selling AI chips; it is now allied with Wall Street to provide money and leverage for data centers, and the data centers that receive the money will return to buy NVIDIA chips. The biggest signal is that they have partnered with financial institutions like BlackRock, Blackstone, KKR, and Goldman Sachs to launch a new $500 billion funding strategy, with institutions funding to build AI factories filled with NVIDIA GPUs. It's like Nvidia, which is pushing funds into the AI industry while turning those funds into its own orders, essentially a circular financing system. If this continues, AI chips and data centers will be packaged as financial assets similar to mortgages or car loans. Investment institutions can first set up dedicated financing tools, borrow money to build data centers, and then use future rent, computing power income, or equipment to repay debts. Although this benefit allows for continued expansion of AI investment, if final demand and revenue fall short of expectations, underlying debt risks may be concentrated. So this guy now sees NVIDIA as the energy giant Enron of the past, which once designed a large number of complex and hard-to-understand financing structures, hid debts in special entities, and packaged energy contracts as tradable financial products to maintain growth and profits. This has led this guy to keep increasing his short positions in semiconductors and memory stocks, with the latest move on August 12. $SOXL [Securitize: Tokenized assets hit new highs, but why are losses actually widening?] 】 RWA leader @Securitize announced its Q2 2026 financial report, with the following key points: ・Average tokenized assets under management reached $4.3 billion, up 16% year-on-year ・On-chain transaction volume grew by 147%, reaching $5.3 billion ・Revenue declined 5% year-over-year, dropping to $14.4 million ・Net loss widened from $6.15 million in the same period last year to $21.7 million Some losses came from fair value changes in options and derivative liabilities, as well as non-cash outflows; Even after exclusion, adjusted EBITDA still turned from profit to loss (1.8 million →-5.5 million), with operating costs up 56%. This indicates rapid growth in RWA adoption and transactions, but Securitize has yet to convert scale into revenue. For the crypto community, this remains an important milestone for institutional assets on-chain. BlackRock's BUIDL has already been implemented through institutional collateral processes, and Securitize has obtained tokenized securities custody and stablecoin atomic settlement qualifications, potentially further linking RWA, stablecoins, and DeFi. Tokens worth noting include $ETH, $SOL, $AVAX, and $ENA related to the USDe ecosystem. Ethereum has a more mature foundation in RWA and DeFi, while Solana and Avalanche continue to take on tokenization funds, equities, and trading infrastructure. However, Securitize itself adopts a multi-chain strategy, and networks like $ARB, $OP, $POL, $APT, $BNB, and $TRX may also disperse related activities.#7月CPI平稳落地, expectations for a rate hike in September cooled Strange: Even with CPI positive, why hasn't BTC risen yet? Last night, the US July CPI came out: Year-on-year growth was 3.4%, down from June's 3.5%; Core CPI 2.5%; It rose only 0.1% month-on-month. The data is not bad; it can even be said to be the data that risk assets like to see. And what happened? US stocks rose, market expectations for a rate hike in September cooled significantly, while BTC was still stalling. On the contrary, I think this is the most noteworthy signal to watch this CPI. When good news arrives but an asset doesn't move, that itself is information. Why? First, the CPI did not "exceed expectations." The market has long known that inflation is likely to cool, so this is not a surprise, just that "nothing has happened." Second, what BTC currently lacks may not be a CPI positive at all, but incremental capital that genuinely wants to push prices upward. Third, tonight there's another card — PPI. If PPI continues to cool, the logic of "no rate hike in September" will be further confirmed. If BTC doesn't move, I will interpret it as a weak signal. Conversely, if tonight's PPI exceeds expectations and BTC still holds its key position, I will actually start to go bullish again. So my judgment is simple: Before tonight, I won't chase BTC. What is truly worth watching is not the "CPI is good but bad is bad," but rather: After the positive news arrives, will money buy BTC or not? Macro data is just the starting gun, Funds must vote with real money—that's the answer. Do you think BTC is just gathering momentum, or has it stopped rising? #BTC #比特币 #CPI #PPI #美联储 #加密货币 #宏观经济 #币圈Corporate Bitcoin Treasuries Are Entering a New Phase Corporate Bitcoin adoption was once built around a simple narrative. Buy Bitcoin. Hold Bitcoin. Never sell. That assumption is beginning to evolve. Strategy recently sold 1,690 BTC, reportedly using part of the proceeds for preferred-share buybacks and cash reserves. At the same time, other corporate treasury strategies continue to diversify—some firms are increasing Bitcoin exposure, others are expanding into Ethereum, while capital management is becoming more dynamic. This represents a natural evolution. As more companies hold digital assets on their balance sheets, treasury management begins to resemble traditional corporate finance. Cash requirements change. Capital structures evolve. Shareholders expect buybacks, liquidity management and balance-sheet flexibility. The question is no longer whether companies will own Bitcoin. It's how actively they'll manage those holdings. Long term, corporate adoption can still create structural demand. But investors may also need to accept that treasury companies won't always be one-way buyers. Corporate crypto ownership is maturing. And mature treasury strategies rarely remain static. Do you think active treasury management strengthens the corporate Bitcoin thesis—or weakens it? Share your thoughts below 👇 #StrategySellsBTCAgain On August 13, the Asian stock market closed against the crypto market The Japanese and Korean stock markets closed first at 14:00, with semiconductor companies as the main weights, directly linked to storage-type crypto mapping coins like $SNDK and $MU. If the Japanese and Korean markets close higher, funds will flow back into stocks, and storage coins will surge; If the market closes lower, domestic South Korean funds will flow into crypto as a safe haven, $BTC find slight support, and small-cap crypto volatility will quickly narrow after the close. The 15:00 A-share close marks the end of domestic capital trading and determines the risk appetite of Asian retail investors. When the market closes down, institutions simultaneously reduce their crypto holdings to recover funds, testing support under pressure $BTC and $ETH. A-shares strengthen, with a small amount of funds allocated to platform coins such as $OKB and $BNB. After the close, the crypto buy-ask spread widens, with no capital support and an increased probability of short-term insertion. The 16:00 Hong Kong market close was a core watershed for the day, with a large concentration of Asia-Pacific crypto institutional funds. A weakening Hong Kong market will trigger institutional margin redemptions, and mainstream currencies are likely to break below the lower boundary of their ranges; If Hong Kong stocks close higher, short-term funds will be lying in wait for crypto gains, awaiting market moves in Europe and the US. After 16:00, Asian trading liquidity will plummet sharply, entering a grinding phase with reduced volume. Singapore's 17:00 close was only for liquidity, with almost no major fluctuations, and the market focus shifted entirely to the European session. In practice, from 14:00 to 16:00, there were frequent shakeouts between bulls and bears, and before the close of Hong Kong stocks, new heavy positions were reduced; When the market weakened across the board, high-volatility counterfeit coins were avoided, with only BTC and ETH serving as short-term safe havens. #芯片股领涨, Korean stocks rebounded over 22% in ten days ⚠️ Market review is only and does not constitute investment advice#特朗普因TruthSocial付费数据流遭起诉 I actually don't care much about "paid data" itself; I care more about one thing: when a single word from one person is enough to cause a huge market fluctuation, the information gap itself becomes a trading advantage. From a business perspective, it's not strange that TruthSocial sells low-latency data to institutions; institutions are willing to pay for speed, essentially buying execution advantage. But the problem is, if the data involves the president's own public statements that might influence the market, then it's no longer just ordinary market data. Some people know in advance by a few seconds or even earlier, and ordinary investors only see the deal afterward; the outcome may be completely different. That's also why I've never liked chasing breaking news. The most likely time to lose money in crypto isn't when you look in the wrong direction, but when you think you're trading news, you're actually taking on someone else's liquidity. Especially now, $BTC $ETH is increasingly susceptible to political, macro, and unexpected events. Speed is important, but I'd rather wait for the first wave of sentiment to release and see if the price has truly held up. Information is responsible for creating volatility, while price is responsible for telling you whether the market believes it.NAND, which has risen tenfold, is it about to start falling? ” Event: Dalian No. 2 Factory has officially resumed operations after four years of shutdown, with plans to move in equipment in November 2026 and production to start in the first half of 2027, with a monthly capacity of 50,000 wafers. Three core impacts: 1. Short-term supply and demand are not affected, but market expectations have changed New capacity will not be released until next year, and supply will remain tight this year. But "expansion" itself is an industry barometer, and funds will trade supply-demand turning points ahead of time. 2. Dual-track division of labor, balancing profit and market share · Dalian focuses on mature processes at 100 levels (inheriting Intel's floating gate technology), emphasizing cost-effectiveness and scale; · The M17 factory in Cheongju, South Korea, focuses on cutting-edge products with 300+ layers, targeting the high-end AI market. This is a typical "defense while attacking" strategy. 3. The logic behind price increases faces long-term challenges Over the past year, NAND prices have risen nearly tenfold, mainly due to the dual impact of "AI demand explosion + supply contraction." Now that supply is easing and capacity is gradually being released in the coming years, combined with downstream inventory cycles, the certainty of sustained price surges is decreasing. Conclusion: Dalian No. 2 Plant is just the beginning; the global capital expenditure cycle for memory chips has officially shifted to expansion. For investors, short-term performance is expected to materialize, while mid-term caution is needed regarding expectations of overcapacity.$OKB OKB is back around the $100 level, and the recent momentum is definitely worth watching. The token has gained more than 16% over the past 7 days, while its total supply has now been fixed at just 21 million tokens. That limited supply is one reason people are once again comparing OKB with BNB. But I wouldn’t call it “the next BNB” yet. The bigger story is how deeply OKX is integrating OKB into the X Layer ecosystem. Combining exchange demand, X Layer adoption, and a fixed 21M supply gives OKB a stronger fundamental narrative than simple hype. Still, there’s an important question: OKB remains well below its previous ATH of $258.6. So buying here isn’t simply betting on another short-term pump. The real bet is whether OKX can make OKB a core asset across its broader ecosystem. If that happens, $100 could eventually look like the beginning. If not, scarcity alone won’t guarantee long-term value. The key thing to watch now is whether OKX keeps adding real utility and demand for OKB. Do you see OKB as an opportunity at these levels, or has it already moved too far? #CPIEasesHikeBets #AIInfraEarningsWatch #SpaceX99%ValueFromAI 一小時熱門榜最容易出現的誤會,是把總量直接當成趨勢。OKX Onchain OS 在 08 月 13 日 14:00 的官方快照顯示,BTC、ETH、SOL 最近一小時分別有 32、22、25 次提及;二十四小時總量則是 1482、638、601 次。 為了讓兩個窗口能比較,可以先把二十四小時總量除以二十四,再用最新一小時去比。結果是 BTC 0.52 倍、ETH 0.83 倍、SOL 1.00 倍。高於一表示最新一小時比全天平均活躍,低於一則表示相對安靜;這只是討論速度,不是報酬率。 按這個口徑,BTC 明顯放慢,ETH 有所放慢,SOL 大致貼近長窗均值。誰的原始提及量最高,未必就是相對自身基線升溫最快的那一個。把「量最多」和「加速最快」分開,能少掉很多誤判。 語氣還要另看一層。BTC 是 偏多略佔優,偏多與偏空分別 34%、22%;ETH 是 偏多明顯佔優,比例為 36%、14%;SOL 則是 偏多明顯佔優,比例為 44%、12%。 這裡的關鍵是分母。ETH 一小時只有 22 次、SOL 25 次,幾條新增文本就可能明顯改變百分比;BTC 雖然樣本較大,也可能包含同一事件的轉發與#CPIEasesHikeBets July U.S. inflation largely matched expectations, with headline CPI easing to 3.4% year-on-year and core inflation slowing to 2.5%. Following the release, market-implied odds of no September rate change reportedly increased to approximately 59.9%. Short-term Treasury yields declined, gold recovered after an initial drop, and Bitcoin remained relatively rangebound. The result weakens the argument for an immediate rate increase without confirming that inflation has fully returned to target. Attention now shifts to PPI and future energy prices. Producer-price pressure could eventually flow into consumer inflation, while fiscal deficits and elevated term premiums may keep longer-term Treasury yields high even if the Fed pauses. My view is that the CPI report is supportive for risk assets, but not strong enough to create a decisive breakout by itself. BTC and equities may need falling real yields and improving liquidity—not merely stable policy expectations—to extend their gains sustainably.$BTC ETF sees 61 million outflows, but $ETH is still attracting funds: Are institutions starting to dislike the big specs? Here comes the most captivating set of data today. Latest disclosure of spot ETF flows on August 12: $BTC saw a net outflow of about $61.1 million, with FBTC outflowing about $46.8 million. $ETH saw a net inflow of about $7.4 million, mainly from ETHA. At this point, your first reaction is definitely: Are institutions starting to switch from BTC to ETH? Let me start with the conclusion: It can't be called a position swap yet, but ETH has indeed gotten a short-term entry ticket better than BTC. Why? First, the scale is still insufficient. BTC outflowed 61 million, while ETH inflowed only 7.4 million. This isn't a "massive capital reshuffle"; it's more like someone on BTC exited first, and ETH still has a buying order yet to leave. The gap is too big; just using a day's data to call an ETH reversal easily fools oneself. Second, the price doesn't match. Within $ETH days, it moved from 1873 to around 1925, and now returns to the 1895 area. If there really is a large amount of funds rushing for ETH, 1900 shouldn't be so hard to stand. Cash flow is good news, but prices have yet to confirm it. Third, the most awkward situation for ETH right now is: some people are buying, but no one wants to chase it. There was indeed support around 1870 below; Above, 1900–1925 were also sell-offs. That's why the market has become like this—neither falling nor rising is smooth. Next, I will look at only two signals: Whether the ETH ETF can maintain continuous net inflows is not just one day in the red; Can ETH climb back to 1900 and truly take over 1925? The simultaneous appearance of these two shows that ETH's relative strength is beginning to be recognized by the market. Otherwise, the $7.4 million inflow would be just a nice topic. Just because someone knocks on the door for an ETF doesn't mean the trend has already entered the room. $BTC $ETH #财报观察员: AI infrastructure financial reports make a succession #三星sk海力士领涨首尔股市 KOSPI breaks into the tech industry, memory chips help Korean stocks recover On August 13, South Korea's KOSPI index rose over 4% in early trading, rebounding about 23% from the July 30 low, officially entering a technical bull market. Samsung Electronics rose over 4%, SK Hynix gained over 7%, with two major storage giants leading the market $SKHYNIX 🔍 Mark Newton, Head of Technical Strategy at Fundstrat, has a clear view: Korean ETFs have broken through key technical levels, and their short-term structure has strengthened. A more crucial signal is that memory chip stocks have started to outperform the overall tech sector again, marking the first time since June. This shows that capital is not being grabbed blindly, but is being targeted to cover the previously worst losses in the storage sector. ⚠️ Marcel Thieliant of Capital Capital Economics warns: South Korea's current semiconductor-driven boom may lose momentum within two years. He predicts that the US AI investment boom will cool down by 2028, at which point Korean chip manufacturers may be forced to cut capital expenditures. SK Hynix did this in 2023—cutting capital expenditure by two-thirds. Samsung and SK Hynix's combined 800 trillion won factory construction plan has yet to be finalized. 👀 A technical bull market is a fact, but it's only a confirmation at the price level, not a get-out card for fundamentals. Memory chips are a highly cyclical industry; when the market rises, it's strong; when the tide falls, it's fast. Short-term momentum is fine; following capital can benefit from it; But don't mistake cycle reversals for growth and sustainability; keep a clear eye when prices rise later. Storage, with Korean stocks making a comeback, is backed by capital during this rebound. But cyclical stocks fear the most about treating the highs as eternal, watching as they play, and not getting carried away. $SNDK Russian Central Bank: Retail investors can trade $BTC $ETH $USDT Russia has opened a door for cryptocurrency, but the crack is not wide. According to a related article by bits.media, a recent draft released by the Russian central bank states that ordinary investors currently have only three types of crypto assets available for future trading: Bitcoin, Ethereum, and USDT. Within a single brokerage, crypto exchanger, or asset management institution, each person's annual purchase limit cannot exceed 300,000 rubles, and a risk test must be passed before trading. Here are some trading restrictions: large market capitalization, high daily trading volume, and at least five years of price records in overseas markets. However, professional investors face relatively relaxed restrictions, can trade other cryptocurrencies, and have no purchase limit, but must meet compliance requirements. This plan is not about Russia fully liberalizing crypto trading, but rather gradually pushing funds that had previously been in the gray area into licensed institutions and regulatory accounts. The relevant system is expected to take effect from September 1, and the Moscow Exchange has already begun preparing its own crypto asset custodian institution. More notably, the Russian central bank ultimately left USDT, a US dollar stablecoin issued by a US company, joining the first batch alongside BTC and ETH, indicating that regulators prioritize liquidity scale over demand. This time, Russia has not accepted the crypto world, because for most altcoins, the door to the compliant market remains tightly closed!