ShizukaKazu

#特斯拉Q3财报 Profits beat expectations, but the stock fell: Tesla’s “divided earnings report” and the ultimate wager on physical AI
An earnings report featuring a gross margin recovery and a stunning reversal in free cash flow should, in theory, have been rewarded by Wall Street. Yet Tesla’s (TSLA) stock was hit by a fierce sell-off after the report. Caption: Tesla (TSLA) intraday trading after its earnings report. The stock initially surged above $395 at the open, but then came under heavy selling pressure, plunged straight down below the previous day’s closing price, and ultimately fell more than 2.4%. This “rise at the open, fall later” movement precisely captures the market’s sharp divide between “profits beating expectations” and “stagnant revenue, surging expenses.”
This is an extremely divided earnings report: profits beat expectations, but revenue showed zero growth, while operating expenses surged nearly 40%. As “traditional automaking tailwinds fade” and the grand narrative of “physical AI” compete, what is the market worried about, and what is it anticipating?
I. A Fading Automaker? “Profit Recovery” Without Regulatory Credits
In total, Tesla’s Q3 revenue was $28.05 billion, essentially flat year over year and slightly below market expectations. But profits delivered a huge surprise: Q3 total gross margin recovered to 19.6%, while automotive gross margin excluding regulatory credits rose to 18.1%, both above expectations. This shows that even as the “profit crutch” of regulatory credits was substantially removed (down 73% year over year), Tesla’s efforts to control manufacturing costs and optimize its regional product mix are translating into genuine earnings quality. However, beneath the veneer of better-than-expected profits, mounting pressure across revenue streams was the direct reason for the capital flight. Automotive revenue fell 3% year over year, while energy generation and storage revenue plunged 16% year over year (far below expectations). More damagingly, to support AI and autonomous-driving research, Tesla’s Q3 operating expenses surged 39% year over year (R&D expenses soared to $2.76 billion). With revenue stagnant, the spending surge directly drove GAAP net income down 43% year over year, while operating margin plunged to 2.6% from 5.8% in the same period last year. This reality weighed on the stock during the session.
II. The Biggest Surprise: A Stunning Reversal in Free Cash Flow
Although the stock was hammered, the earnings report still contained a highly meaningful bright spot. In today’s high-interest-rate macro environment, Wall Street is far more sensitive to “cash flow” than profits. Analysts had widely expected Tesla’s Q3 free cash flow to be negative $4.99 billion. But after the earnings report was released, Tesla’s free cash flow actually turned positive, reaching $1.53 billion. Operating cash flow of $6.26 billion fully covered capital expenditures of $4.73 billion, driving cash and short-term investments at quarter-end up to $45.45 billion. This stands in stark contrast to our earlier article, “Amazon’s Chips Gather Dust Without Power”: while other tech giants turn free cash flow negative by blindly expanding AI infrastructure, Tesla has demonstrated through its exceptionally strong ability to generate cash internally that it has ample means to support its substantial AI capital spending ahead.
III. The AI Money Printer: FSD and Robotaxi on a Tear
What supports Tesla’s 358x price-to-earnings ratio is its long-term AI and autonomous-driving narrative, and this quarter, those narratives are becoming reality:
FSD (Full Self-Driving): Active subscriptions reached 1.74 million, nearly 11% above expectations; the take rate on new vehicle deliveries in North America exceeded 75%. This means software subscriptions are becoming Tesla’s core high-margin money printer.
Robotaxi (self-driving taxis): Unsupervised service using Cybercab began in Austin in September. By quarter-end, cumulative unsupervised miles driven were close to 2 million, up 8x quarter over quarter. FSD provides the driving capability, while Robotaxi handles commercial operations. Tesla is no longer simply selling cars; it is building a vast autonomous mobility network.
IV. The Ultimate Trump Card: 400MW of Supercomputing Power and Optimus Mass Production
If there is a strongest foundation for Tesla’s valuation, it is that the company is becoming a “physical AI infrastructure” company. The earnings report disclosed that Tesla plans to reach 400MW of Superintelligence (SI) training compute capacity by the end of this year. The Cortex 1 and Cortex 2 clusters in Texas are both operational, while equipment commissioning at the Austin Fab is planned to begin in 2027. This vast computing capacity will serve not only FSD’s development, but also the Optimus humanoid robot. The report shows that Optimus has begun production on a high-capacity production line in Fremont, California, and the robots coming off the line are the first manufactured by the company to meet mass-production quality standards. The cost is also very real: Q3 R&D expenses soared to $2.761 billion, and operating expenses rose 39% year over year. But this is Tesla’s calculated strategy: trade short-term profits for the ultimate ticket to the physical AI era.
Conclusion: Whom Should We Believe?
Tesla’s Q3 earnings report gives both bulls and bears reasons to make their case—a truly “divided earnings report.” Bears are focused on zero revenue growth, the energy business missing expectations, GAAP net income being halved, and surging operating expenses, and are selling the stock on the belief that this “cash cow” is being drained by heavy AI capital spending and price wars. Bulls, meanwhile, are fixated on free cash flow turning positive, surging FSD subscription rates, and 400MW of supercomputing power and Optimus mass production, convinced that Tesla is no longer just a carmaker. The 358x price-to-earnings ratio the market assigns Tesla has never been based on it being an “automaker.”
While traditional automakers are still locked in price wars to sell cars, Tesla is trading short-term profit pain for the ultimate ticket to the physical AI era. The short-term stock decline is simply the market using a calculator to check the current bill; Tesla’s real trump card lies in the supercomputing clusters that are about to come fully online.

#Meta财报超预期 I care more about future guidance, especially capital expenditures and the path to AI returns.
Revenue shows how much a company sold over the past quarter, earnings show how much it made during that period, while guidance reflects where the company plans to spend and how quickly its business can grow.
For companies like Meta and Tesla undergoing AI transformations, current-period figures can easily be distorted by one-off expenses or the pace of R&D. Meta’s key lies in Muse and returns on computing power. Q3 revenue grew 26%, and fourth-quarter revenue guidance was also above expectations, showing that its core advertising business remains strong. But the full-year capital expenditure ceiling is still $145 billion, and free cash flow has already turned negative.
The real question is whether these investments in computing power can translate into Muse users, transaction revenue, and profits. More than 10 million people in North America have already created personal AI agents, and about a quarter use them daily, but global scale and steady profitability are still some way off. Tesla’s key lies in how quickly it can deliver on Robotaxi and Optimus. Q3 revenue was nearly flat, automotive revenue fell 3% year over year, and GAAP net income fell 43%. Turning free cash flow positive is certainly encouraging, but some of that came from slower capital spending. Full-year capital expenditures were lowered from more than $25 billion to about $20 billion, prompting market concerns that projects like Robotaxi and Optimus may be delayed.
The signals from the earnings call were also somewhat cautious: robot deliveries to external customers are still about a year away, while Robotaxi is not expected to make a meaningful contribution to profits until the second quarter of next year.
So I would break this earnings report down into two points: Meta proved it can still grow rapidly, but AI investment is eating into profits and cash flow; Tesla proved it still has cash resilience, but its core auto business is weak, and AI and Robotaxi have yet to generate new profits. If I had to focus on just one thing, future guidance matters more—it directly answers one question: How much return will the money a company spends generate in the future?#WCTCS9

#SpaceX市值超越台積電 SPCX still has a sentiment premium in the short term, but further significant valuation upside requires continued fulfillment of AI computing-power orders and improved cash flow.
This looks more like a “post-earnings revaluation” than a move driven solely by surpassing TSMC in market capitalization. At the October 21 close, SPCX stood at $159.31, up 7.79% on the day, with a total market capitalization of about $2.16 trillion. The share price has already priced in the positive third-quarter results[9].
The results were indeed strong: third-quarter revenue was about $14.6 billion, up 161% year over year; net profit was $4.4 billion, compared with a loss of $1.1 billion in the same period last year. The company achieved its first quarterly profit, an important turning point in its valuation story.
Growth was mainly driven by two factors:
AI computing power became the largest source of revenue: the Superintelligence business generated $7.56 billion in revenue, up 830% year over year, accounting for about 52%.
Starlink remains the main source of profits: the connectivity business contributed most of operating profit, while government, enterprise, and aviation customers expanded rapidly.
SPCX’s story has therefore changed: it is shifting from a “space launch company” to a combination of “AI computing power + satellite network.”
Can the valuation rise further? Yes, but the bar is high!
The bullish case is clear: AI cloud contracts continue to convert into recognized revenue, computing capacity maintains high utilization, and the additional capacity from fourth-quarter capital spending is monetized quickly. In that case, the market will view SPCX as a scarce AI infrastructure asset, and its valuation multiple may remain high. The risks are just as clear.
Third-quarter capital spending was about $19.6 billion, most of it directed toward AI infrastructure; free cash flow remains negative. Fourth-quarter capital spending is expected to rise further to $35 billion–$40 billion, indicating that expansion is still accelerating.
This means a key test lies ahead: can new revenue keep pace with new investment? If AI revenue growth slows, or cash flow improves more slowly than expected, the current high valuation will come under pressure. There is another practical detail: SPCX’s P/E ratio is still negative, and its share price is supported mainly by growth expectations and the AI narrative, rather than the stable profits of a mature software company.
The bond market is also reflecting this divergence—after the results were announced, SpaceX bond spreads widened, indicating that equity investors are focused on growth while creditors are more concerned about financing and debt risks.
Overall, SPCX is likely to remain strong at elevated levels in the short term, especially if it continues to land large AI contracts, increase computing-capacity utilization, or improve free cash flow. Whether it can truly break into a new valuation range over the medium term will mainly depend on three factors:
1. Whether AI revenue can repeatedly exceed expectations;
2. Whether Starlink can continue to provide stable profits;
3. Whether massive capital spending can translate into positive free cash flow.
SPCX has already completed one round of valuation expansion on the back of its earnings report. Whether it can climb another step in the next phase will depend on how quickly its computing-power investments generate returns, rather than short-term changes in market-cap rankings.#WCTCS9 $SPCX ‌

#特斯拉Q3利润同比下降43% Tesla is in a transition phase of “automotive as a floor, AI driving valuation.” Short-term profits are under pressure, but cash flow and orders have not lost momentum; whether its valuation can continue to rise in the medium term depends on whether Robotaxi and FSD can turn technological progress into replicable profits.
Delivery pressure comes from supply, not a collapse in demand. Q3 deliveries were 486,500, down 2% year over year; production was 464,400, up 4%. The passenger vehicle order backlog at quarter-end reached a four-year high, and the company said deliveries were mainly constrained by production capacity and battery supply.
This shows that there are still buyers in the market. The pressure comes from price cuts squeezing automotive revenue, lower regulatory credit revenue, and continued increases in expenses from R&D and computing investment.
What truly supports the valuation is that the AI business is becoming verifiable: FSD
FSD take-up on new vehicles in North America exceeded 75%, global paid users topped 1.7 million, and subscriptions accounted for more than half for the first time. Robotaxi’s cumulative unsupervised miles grew 8x, while the Cybercab fleet expanded about 10x in roughly a month.
These figures show that autonomous driving has moved beyond the purely conceptual stage. But the Robotaxi cost target of 30 cents per mile has yet to be achieved, and standalone revenue and profits have not been fully disclosed.
Lower capital expenditure has eased short-term cash flow pressure.
The full-year capex guidance was cut from more than $25 billion to about $20 billion, Q3 free cash flow turned positive at $1.531 billion, and cash and investment balances exceeded $45 billion[5]. This looks more like pushing back the pace of some investment, not exiting expansion. Management said explicitly that it will continue to increase investment in Robotaxi, Optimus, computing, and semiconductors over the coming years. Three things to watch next:
Whether Robotaxi can be replicated across cities: Austin is just the starting point; operating performance in Dallas and more cities will be more informative.
When unit economics will improve: Vehicle utilization, cost per mile, and gross margin will determine whether the business can generate real profits.
Whether the automotive core can stop declining: No matter how large the AI option may be, automotive, energy, and services still need to provide sustained cash flow.
Overall, Tesla has moved beyond the stage of simply talking about “the number of cars sold” and entered a period of validating AI commercialization. In the short term, the stock price remains exposed to automotive-cycle fluctuations; whether the valuation can continue to expand in the medium term depends on whether Robotaxi delivers revenue and profits in 2027.$TSLA ‌

#Solana上季非投票交易达89.2亿笔 This growth has a real foundation!
In Q3 2026, Solana recorded 8.92 billion non-vote transactions, up 104% from Q4 2025. Non-vote transactions more closely reflect actual usage, such as user transfers and trading, and are a better indicator of on-chain activity than total transaction volume.
Application revenue also rebounded, rising 48% quarter over quarter to $299 million. Quarterly spot DEX trading volume was approximately $197.2 billion, accounting for 36% of cross-chain spot trading volume and remaining above 30% for nine consecutive quarters.
Tokenized stocks were another bright spot, with quarterly trading volume reaching $7.1 billion, including $4.4 billion in September alone, indicating that Solana is capturing demand for high-frequency, small-value settlements.
Four factors will determine whether high growth can continue
Can transaction volume translate into revenue? A large number of small transactions can drive up transaction counts, but fees are extremely low. Protocol revenue and cash flow are what truly determine the ecosystem's momentum.
Are funds willing to stay? TVL represents the amount of capital locked in DeFi. Complete TVL data aligned with transaction counts is currently lacking, so this needs to be monitored.
Can the user experience remain stable? The Votor upgrade eliminated on-chain voting transactions, making non-vote transactions a better measure of network load. One hour after the upgrade, the rate was approximately 2,500 transactions per second, broadly in line with previous levels.
Will market conditions be supportive? BTC and ETH price trends, the price of SOL, and financing conditions will continue to affect new project launches and user participation.
My view
Solana has established a strong advantage at the transaction layer, with room for further growth in payments, DEXs, stablecoins, and tokenized assets. It could sustain medium-to-high growth in the next phase. But it is too early to judge the entire ecosystem as overheated based solely on “8.92 billion transactions.”
A more accurate conclusion is that transaction activity is leading and revenue is recovering; whether high growth can be sustained depends on capital retention and actual returns.#WCTCS9 $SOL ‌

#PHP严重漏洞引发政府网站下线 The PHP vulnerability crisis has triggered a global emergency “pull-the-plug” response—don’t count on AI to save the internet!
What exactly happened?
PHP is the underlying language for a large number of websites, government systems, and content management systems. CVE-2026-109618 and CVE-2026-109619 have been disclosed; both vulnerabilities have a score of 9.2 and are high-risk remote code execution vulnerabilities. Simply put, attackers don’t need to log in: they can send malicious requests to vulnerable servers and potentially gain control of the system, read memory data, or steal credentials.
The vulnerability details and exploit code are now public, and “in-the-wild exploitation” has emerged, meaning attacks are already taking place. The official fixes were released on October 21, 2026:
· PHP 8.2 → 8.2.35
· PHP 8.3 → 8.3.36
· PHP 8.4 → 8.4.27
· PHP 8.5 → 8.5.128.1 and earlier versions have ended standard support, and the official team no longer provides formal security releases.
Will the impact continue to grow?
Yes, mainly due to three types of lagging scenarios:
Legacy systems: PHP websites, backends, and admin panels that have not been upgraded for a long time.
Unmaintained assets: test sites, official websites, and third-party service provider systems.
Processes not restarted after an upgrade: if only the program files are replaced while old worker processes continue handling requests, the risk remains.
As a result, the affected scope could continue to spread from government websites in Europe and the US to corporate websites, e-commerce platforms, hosting platforms, and cloud services. The impact will be more pronounced on systems that actually suffer data leaks or server takeovers.
What does this mean for the crypto market?
The impact is concentrated on exchanges, wallet service providers, project websites, and off-chain backends—not on the blockchains themselves:
Short-term sentiment is negative: the market may worry about attacks on platforms such as exchanges and wallets, potentially triggering a brief flight to safety.
Some platforms may be affected: SlowMist investigated 1488 crypto-related domains and found that 793 had clear PHP fingerprints, indicating that some platforms do use the relevant tech stack. Coinb and BitG have both said their platforms are unaffected.
User-side risks are more tangible: hackers may forge login pages or impersonate patch links to trick users into handing over account details, API keys, or seed phrases.
On-chain assets will not be affected automatically: public blockchain networks such as Bitcoin, Ethereum, and Solana continue to operate as usual, and private keys in wallets under your control will not disappear automatically because of a PHP vulnerability.
What should we expect next?
The vulnerability’s impact will likely continue to emerge for some time, until websites worldwide have completed upgrades and checks. The overall impact on the crypto market is limited, but project websites, smaller exchanges, and wallet services with weak security awareness may face greater pressure.
For ordinary users, the priority in the near term is to verify platform announcements, access platforms through official channels, and beware of links for “urgent patches” or “airdrop verification.” In terms of asset prices, this looks more like an external security incident than a direct attack on crypto protocols themselves.
AI’s contribution to this incident deserves recognition: with AI, researchers can continue tracing leads, analyzing why attempts failed, and advancing work that was difficult to accomplish in the past.
But proving that a vulnerability can be exploited often means producing a working set of attack code. The models’ analytical and debugging capabilities demonstrated in this process could also be used to improve attack tools.
As the barrier to this kind of work falls, the threats maintainers have to deal with will also grow.
This time, researchers submitted their reports in advance, buying time for a fix. How much preparation time will be left in the future, however, is difficult to predict.
AI can find and verify problems ever faster, but whether the entire internet can keep up with the pace of patching is another matter.

#Gate第三季度报告出炉 Gate’s Q3 report card is out—don’t just focus on trading volume; there’s more to it than you think.
From crypto trading to global stocks and ETFs, then wealth management and payments—Gate’s financial footprint keeps expanding.
Here are a few key numbers:
🏆 Derivatives trading volume hit $396 billion in September, surging 38.2% month over month and putting Gate straight into the global top three—this is no small feat; there’s real volume behind it.
📈 Coverage includes 12,800+ global stocks and ETFs. Whether you want to trade US stocks, ETFs, or multiple asset classes, you can do it all on one platform.
💰 YuBiBao’s assets surpassed $1.85 billion, up more than 24% in a single quarter—the wallet earns yield passively and just keeps growing.
From trading volume to asset management and comprehensive finance, Gate is making good on the promise of an “all-in-one financial ecosystem.”
If I could bet on just one direction, I’d pick global stocks and ETFs.
There’s just one reason: crypto trading is already the foundation (derivatives are already in the global top three). The most exciting growth opportunity ahead is turning crypto users like us into “all-asset players”—being able to trade crypto and buy US stocks and ETFs directly from one account. That’s the real value of “all-in-one finance,” and the strongest way to keep users on the platform.
YuBiBao is the steady, quietly profitable route, and its growth is impressive (+24%), but when it comes to “breaking new ground,” I’m more excited to see the asset footprint expand outward.

#Gate30日净流入全球第二 On October 21, 2026, South Korean stocks opened sharply higher, with the KOSPI Index surging 2.05% at the open before extending its gains to 3.83%. The key driver was a broad-based breakout in semiconductor and memory chip stocks.
1. Market performance: The two memory chip giants led gains
· Index and leaders: As of publication, SK Hynix surged 7.52% and Samsung Electronics rose 6.15%, with the two memory giants directly driving the broader market higher[11].
· Sector-wide gains: South Korean semiconductor stocks gained 4.66% overall, while chip materials, packaging and testing, and equipment stocks across the supply chain rose in tandem, creating clear buying momentum.
2. What sparked the rally: An oversold rebound and support from overseas markets
· Improved sentiment in U.S. stocks overnight: U.S.-listed hardware stocks, including Micron, held firm overnight, boosting confidence among Asia-Pacific tech investors and drawing foreign capital back into memory stocks.
· Earlier oversold conditions and technical buying: Amid concentrated selling by foreign investors, South Korean stocks had pulled back for several consecutive sessions, leaving leaders such as SK Hynix deeply oversold. As valuations quickly adjusted, technical buyers entered to bet on a rebound.
· Expectations for the memory cycle: Investors are still weighing the sustained demand for HBM (high-bandwidth memory) and storage chips from AI servers, strengthening the appetite to buy on dips[8].
3. How sustainable is the rally?
· Buyback support is fading: Samsung Electronics’ buyback plan, which had supported the market, has been completed, while SK Hynix’s buyback is nearing its end. Their role as significant buyers is weakening.
· Foreign outflow pressure remains: In the previous trading session, foreign investors recorded net sales of 636.7 billion won in South Korea’s securities market, while institutions also posted net sales of more than 460 billion won, leaving clear short-term overhead from trapped positions.
Summary: Today’s sharp rise in chip stocks was mainly a recovery rally driven by an oversold rebound and improved sentiment in U.S. stocks. If the two memory leaders continue to climb on strong volume after the open, the rebound trend may continue; if they quickly retreat after opening higher, it would indicate weak buying conviction and could mean the move is only a short-term spike.

#Gate闲钱宝规模突破3亿美元 Gate Idle Wealth surpasses $300 million—stablecoins earn interest while sitting idle, balancing trading and investing

How do you usually manage the USDT/USDC you have on hand?
Is it sitting in your account waiting for market moves, or do you put it into an investment product? The biggest dilemma is—put your money into an investment product and you can't trade when an opportunity comes along; leave it available for trading and it just sits idle.
Gate Idle Wealth has just surpassed $300 million in total holdings, solving this very problem:
📊 Current indicative annualized yields
- USDT: up to 5.00%
​- USDC: 1.50%
​- USD1: 6.50% base rate, up to 8.16% if eligible
💡 Key advantage: no subscription required, no lockup
Once enabled, eligible stablecoin balances in your account automatically earn yield. The money is still yours and can be used for trading anytime. No need to keep jumping between investing and trading.
A total of $300 million shows that many people have already voted with their feet—after all, idle funds have a cost too, and earning a little is better than nothing.
Stablecoins sitting idle anyway? Here's my 3-layer capital allocation strategy
After trading for a while, I noticed a common problem—when the market is slow, USDT just sits in my account, and seeing the balance stay unchanged gets frustrating.
Over time, I gradually worked out a "3-layer allocation" approach. Sharing it with everyone:
Layer 1: Immediately available trading funds (30%-40%)
Keep enough funds ready to enter the market at any time, in your spot account.
This money is for seizing opportunities; when the market moves, you can act immediately without hesitation.
Key point: These funds must not be locked up and must be available to withdraw and use anytime.
Layer 2: Idle funds earning yield (40%-50%)
Put stablecoins you don't need for now into Gate Idle Wealth to earn yield automatically.
No manual subscription or lockup required; your balance automatically earns yield.
USDT currently offers up to 5% annualized, while USD1 can reach around 8%—
On 10,000 U, that's tens of dollars a month just for letting it sit—enough to buy a cup of coffee.
Key point: When you need the funds, you can use them for trading immediately without any hassle.
Layer 3: Medium- to long-term investments (10%-20%)
This is your real "long-term money." You can put some of it into products with longer lockup periods
in exchange for higher yields. But the condition is—you know you won't need this money for a while.
💡 Core idea:
It's not a choice between "all-in waiting for the market" or "all-in investing,"
but rather dividing your money into layers, each with its own purpose.
Trading funds stay flexible, idle funds earn yield, and long-term funds pursue higher returns.

Gate Idle Wealth has just surpassed $300 million in total assets, showing that more and more people are starting to value the "efficiency of idle funds." After all, not losing money in a bear market is already a win; earning a little on idle funds is an even bigger win.

💬 How do you usually allocate your stablecoins?
Are you an all-in-waiting-for-the-market type, or do you allocate in layers?

#SK海力士涨超7% South Korean stocks surge to a “circuit breaker”! How long can this rally last?
The sustainability of this rally in Korean and chip stocks is in doubt. It is more likely a short-term oversold bounce than the start of a new long-term bull run. Whether it can turn into a genuine reversal will depend on the results of leading companies going forward.
The rebound came fast: On October 21, South Korea’s KOSPI rose nearly 4.8% at one point, while SK hynix and Samsung Electronics both gained more than 7%. Intraday gains briefly widened to around 7.5%, showing how aggressively investors chased the rally in the morning.
The foundations of the decline remain: SK hynix’s U.S.-traded shares fell about 11% over the past five trading days and about 28.5% over the past 20 days. Its shares on the Seoul market have also fallen about 57% from their 52-week high, so today looks more like a recovery after a sharp drop[1].
The catalysts were largely sentiment-driven: Super Micro Computer issued a strong earnings outlook, easing concerns about AI spending; a rebound in the Philadelphia Semiconductor Index also lifted memory stocks across the board.
Fundamentals still provide support: Citi expects SK hynix’s third-quarter operating profit to rise 34% quarter over quarter and 614% year over year, and believes tight memory supply could persist through 2028. This is the main basis for bullish investors.
Buyback support is fading: Corporate buybacks had previously been an important source of demand supporting Korean stocks, but Samsung Electronics has completed its buyback, while SK hynix’s is also nearing its end. Foreign investors and institutions are still net sellers, so it remains to be seen whether buyers will step in at higher prices.
Watch three signals next:
1. Can SK hynix hold around 1.3 million won and regain the 1.4 million won level? Only then would the rebound be considered effective.
2. Can Samsung Electronics rise in tandem? A sharp gain in a single stock can easily reverse; only if both leading stocks hold steady would it signal a strengthening sector.
3. How U.S. memory stocks perform tonight. The moves in Micron, Western Digital, and others will directly affect sentiment among Korean investors tomorrow.
Korean stocks could extend this short-term rally for another day or two, but consecutive surges will be hard to sustain; a choppy recovery lasting several days to a week, followed by divergence, is more likely. If earnings deliver and U.S. stocks continue to cooperate, chip stocks may have a chance to turn this rebound into a medium-term rally. #Gate30日净流入全球第二

#Gate发布9月透明度报告 I’ve been using Gate for so long—checking in to celebrate the 60 million milestone! Let’s talk about the feature I can’t live without.
When I saw Gate’s September transparency report, showing that its user base had surpassed 60 million, my first thought was: time flies.
From opening an account to trade on a friend’s recommendation to now opening Gate every day and checking Gate Square before I even look at the charts, this is one feature I really can’t live without.
Why do I use Gate Square the most? Here’s what my daily routine looks like:
🌅 In the morning: Check Square for overnight market updates
The first thing I do when I wake up isn’t check prices—it’s see what everyone was talking about on Square last night. Which coin suddenly surged, which project made an announcement, whether there was any breaking news—five minutes on Square is much more efficient than scrolling through a dozen groups.
💬 At lunchtime: Check out everyone’s trade screenshots and chats
I scroll through during lunch. Someone shares a profitable trade and spreads the good vibes; someone complains about getting wicked and I can relate; sometimes I even see a newbie asking a question, with a bunch of experienced traders eagerly answering below. It doesn’t feel like being on an exchange—it feels like joining a group of old crypto friends.
📊 In the afternoon: Dig into real feedback on projects on Square
When I come across a new project I want to learn about, I search for keywords on Square first—it’s much more genuine than the project’s official white paper. Some people share hands-on experiences, others point out the downsides, and some compare returns. You can see both sides, which helps me make decisions with confidence.
🌙 In the evening: Review how everyone traded today
At the end of the day, I browse Square to see whether the pros took profits or added to their positions, and whether futures traders got liquidated or stopped out. Sometimes, when I’m unsure about a trade, looking at everyone’s discussions and ideas helps me get some clarity.
My favorite thing about Square: It feels real
It’s not just a wall of marketing accounts spamming posts—people from all walks of life really do share things:
- Some share their gains, and others share their losses
- Some are bullish, and others are bearish
- Experienced traders share strategies, and newbies ask questions and learn
- Project teams occasionally host AMAs and answer questions
There’s a lot of variety, but it’s valuable variety. Much more interesting than places where everyone shares the same opinion.
Speaking of the September transparency report
Putting the data out there for everyone to see each month—user growth, asset information, operational data, all laid out clearly—is a show of confidence in itself. 60 million users didn’t accumulate in a day; keeping them around takes both a great product experience and transparency.
I hope Square’s content keeps getting richer, and I hope the reports get more detailed. As a longtime user, I’m feeling confident.
💬What’s the first thing you do when you open Gate every day?
Do you check the market, place trades, or, like me, browse Square first?#Gate30日净流入全球第二

#Gate最新储备金率达118.49% I just finished reading Gate’s latest reserve report, and my first thought was: when choosing a crypto trading platform, it’s never about whose marketing campaigns are flashiest. “Tangible security metrics” are what give you real peace of mind.
The report lays the platform’s finances bare: total reserves reached $8.688 billion, up a full 5.7% from the May report, while the overall reserve ratio climbed to 118.49%, far exceeding the 100% asset coverage benchmark.
When I choose a trading platform, I never look at just one number. I keep a close eye on several core security dimensions:
First, the individual reserve ratios for major assets like BTC and ETH. This time, Gate’s BTC reserve ratio was 121.6% and its ETH reserve ratio was 123.78%. Overcollateralization of major assets is the first line of security for users with large holdings;
Second, the completeness of coverage across all coins. This report covered a full 500 crypto assets—not just a few top coins as window dressing. Even obscure small-cap coins have corresponding reserve mappings that can be checked;
Finally, you have to check whether users can verify things themselves. With the Merkle tree and zero-knowledge proof tools included in the report, even new users can follow a few steps to confirm that their account assets are fully included in the reserve pool, instead of having to rely on trust alone.
I’ve seen far too many platforms in crypto use high cashback and new-user perks to grab attention, only to quietly disappear later. It’s reassuring to keep every bit of your assets on a platform with fully transparent metrics.
Curious to hear from everyone here: when choosing a crypto trading platform, which hard security metrics do you scrutinize most? Share your criteria in the comments, and let’s help each other steer clear of trouble～.

#以太坊站上2450美元 Ethereum price defies 15 consecutive ETF outflows near $2,400: Is $2,548 the next target?
Ethereum price prediction: Technical support holds near $2,324 to $2,362
The ETH/USD daily chart shows Ethereum has been consolidating over the past few days. On the daily chart, the 100-day exponential moving average at $2,362 closely aligns with horizontal support at $2,324. Ethereum has tested this area multiple times but has failed to close below it, helping buyers push the price above $2,400.
Holding this area would preserve an important technical support line. A daily close decisively below this support would weaken the rebound structure and put the on-chain cost-basis support at $2,311 further in focus. Further downside could reach the next technical support at $2,215.
Ethereum’s first significant upside resistance levels are the 50-day EMA at $2,444 and the 20-day EMA at $2,500. These averages have capped the recent rebound. A sustained break above both would improve the short-term outlook and could pave the way for a test of horizontal resistance at $2,548.
Additionally, $2,781 presents higher resistance. The Relative Strength Index and stochastic indicator have both turned upward but remain below neutral levels. This suggests bearish momentum is weakening, although the bulls have yet to establish a clear advantage.
For now, holding the $2,324 to $2,362 range would help sustain the rebound, while reclaiming $2,444 to $2,500 would provide stronger evidence of a continued recovery.#WCTCS9

#ZEC暴涨8.5% ZEC is indeed very strong in the short term, but it looks more like a leader in the privacy sector; for now, there’s no clear sign it can single-handedly drive the entire sector sharply higher.
As of October 21 at 17:06, ZEC was trading at $1,267.71, after reaching an intraday high of 8,589.30 yuan (about $1,286), with sharp volatility.
Why this rally is strong:
Event catalyst: The community just completed its largest vote ever, with 2.4 million ZEC participating. The vote supported continuing block reward halvings by a margin of 106:1. Deflation expectations directly fueled buying[.
Capital is rotating: The market is no longer focused only on BTC/ETH and has started looking for niche sectors. Privacy coins have a compelling narrative, ETFs, and real usage to support them.
Sector leader status: ZEC accounts for about 62% of the privacy sector’s market cap, and when it moves, XMR, DASH, and others often follow.
What comes next:
Conditions for continued strength: Hold the 1223–1250 range and break through 1320–1400 on rising volume. Only if it holds above 1400 could it have a chance to challenge the previous high near 1700 again.
Signs of a sector-wide move: XMR returning above $600 and DASH also starting to catch up would indicate a genuine sector effect.
Risks to watch: This rally is partly event-driven, and prices could easily retreat once the positive news is priced in. In addition, selling volume in the futures market has a slight edge, and aggressive spot selling pressure is also significant, so chasing the rally carries considerable risk.
Therefore, ZEC could remain strong in the short term, but whether it can lift privacy coins depends on whether it can break through 1320–1400. If it can’t, the leader will rally on its own; if it can, the whole sector may heat up.
⚠️ Coins that surge like this are highly volatile, so positions must be managed carefully to guard against the risks of false breakouts and rapid pullbacks. #WCTCS9 $ZEC ‌

#美伊斡旋消息打压油价 A Single Headline Triggered a Deep V-Shaped Rebound: Three Scenarios for Crude Oil
During the overnight session on October 20, international crude oil saw a sharp, classic “headline-driven” swing. Brent crude briefly surged to a five-month high of $111.95/barrel, while WTI crude rose to $101.37/barrel. Then a geopolitical headline hit, sending prices plunging: Brent touched a low of $105.95/barrel, and WTI fell to $95.35/barrel, with the intraday decline briefly widening to around 3%. But the sell-off did not continue; prices quickly rebounded from their lows. At settlement on October 21, WTI closed at $101.79/barrel and Brent at $108.79/barrel. SC fell 4.47% in the overnight session.
The session left behind not a one-way trend, but a huge star candle—and a more pressing question: what exactly was the crude oil market trading that night?
The immediate trigger for the decline was a rumor about US-Iran talks.
Media reports said Qatar and Pakistan had submitted a compromise proposal to the US and Iran, which had received “preliminary approval.”
The market’s first reaction was direct: if the talks made a breakthrough, shipping restrictions in the Strait of Hormuz could gradually be lifted, marginally improving the global crude supply outlook. The key factor that had been driving oil prices higher was the risk premium stemming from geopolitical conflict in the Middle East; once a “sign of easing” emerged in the source of that premium, long positions were closed en masse during the thinly traded overnight session, rapidly amplifying the decline. Meanwhile, the resumption of Russian diesel exports was an overlooked marginal headwind. As the US suspended some sanctions and moved to lift export restrictions, at least two tankers carrying around 300,000 barrels of diesel had set sail for Turkey and Brazil, respectively. The marginal recovery in refined-product supply weakened the upward support that the diesel market had previously provided to crude valuations. This signal and the US-Iran news resonated, together providing the fundamental backdrop for the decline.
However, the rebound was almost as rapid as the drop.
The first force behind the rebound was the swift reversal of the news itself. Diplomats involved in US-Iran contacts promptly denied that an “agreement was imminent,” stressing that talks were ongoing but that claims an agreement was close to being finalized were inaccurate. Expectations shifted in the space of a few hours from “peace is within reach” to “it is too early to say.”
The deeper force behind the rebound was renewed recognition that tail risks had not disappeared because of diplomatic rumors.
On the same day, another maritime security incident was reported near the Strait of Hormuz. Iran’s Revolutionary Guard claimed that two oil tankers passing without authorization had struck mines and caught fire; the Houthi threat to attack Saudi oil facilities also remained.
Iran’s conditions for reopening the strait include lifting the naval blockade, unfreezing assets, removing sanctions on oil exports, and withdrawing US forces from the border, among seven key demands. This means that even if a negotiating track exists, there is still a considerable distance to go before any meaningful easing.
After the decline, the market recognized anew that the “source” of the risk premium could contract temporarily on a single headline, but its “foundation”—the tail risk of an actual supply disruption—had not fallen by a corresponding amount.
In addition, API inventory data provided fundamental support: for the week ended October 16, US API crude inventories fell by 4.931 million barrels, while gasoline and distillate inventories also declined, by more than expected. The Strategic Petroleum Reserve, already at an extremely low level, fell to 282.7 million barrels, its lowest since 1982, further magnifying the marginal impact of any supply disruption on prices.
Looking ahead, the core tension in the crude oil market can be summed up as a tug-of-war between downward pressure from the directional unwinding of the geopolitical risk premium and the fundamental support of tail risks to supply disruption.
In the base-case scenario, oil prices are more likely to remain high and trade in a wide range than to continue rising in one direction or fall in a sustained trend. If verifiable progress in talks emerges, the roughly $10 to $15/barrel geopolitical premium currently priced in will face concentrated unwinding, and the average Brent price for the fourth quarter could fall to the $103 to $108/barrel range. But the complexity of Iran’s seven conditions means the premium is more likely to unwind in “steps” than in a “waterfall.”
If talks remain deadlocked and tail risks materialize intermittently, a wide-ranging oscillation between $105 and $112/barrel for Brent is the most probable path. Actual supply vulnerabilities and extremely low inventories would provide a floor, while downward revisions to demand expectations would cap prices; around $120/barrel could be a potential ceiling for a short-term, sentiment-driven spike.
If shipping through the Strait of Hormuz is substantially disrupted, or key Saudi facilities suffer quantifiable production capacity damage, oil prices will see a sharp upward spike. But if this scenario is triggered, it will also activate demand destruction, strategic reserve releases, and stronger incentives for diplomatic intervention, potentially making the spike shorter-lived than a linear extrapolation would suggest.
For crude-related assets, the defining features of the current environment are not a clear trend, but persistently high volatility and the unpredictability of event-driven pricing. The October 20 “sell-off and rebound” is a concentrated example of this dynamic: an unverified headline can trigger a 3% decline within hours, and its refutation can reverse the move just as quickly.
In this environment, the risk-reward profile of one-way directional bets is asymmetric. A more sensible allocation approach should prioritize risk hedging over return enhancement. The rationale for higher valuations in upstream oil and gas assets remains intact, but investors should leave a safety buffer for the potential stepwise unwinding of the geopolitical premium. Downstream chemicals, meanwhile, face pressure from both high cost-side volatility and downward revisions to demand, and are less able than crude itself to keep pace with rising prices.
The signals to monitor most closely, in order of priority, are verifiable changes in actual traffic through the Strait of Hormuz; verifiable progress on specific terms in US-Iran talks; actual OPEC+ production and export data; and marginal trends in US inventories and refinery margins.

#CBOE扩大期权交易时段 CBOE’s expanded options trading hours will significantly boost market activity over the medium term, with limited incremental gains in the short term
I. Key factors supporting increased activity
1. Aligning stock and options trading hours
This fully resolves the previous mismatch in which stocks could be traded pre-market and after-hours, but options could not be traded at the same time, allowing investors to respond immediately to unexpected news.
2. Unlocking incremental global capital
European and Asian investors will be able to trade U.S. stock options directly during local trading hours without staying up all night, substantially expanding the pool of participants.
3. Existing early data confirms the trend
In the third quarter of 2026, average daily options volume during CBOE’s pre-market session reached a record 228k contracts, with a clear upward trend.
II. Practical constraints on near-term activity gains
1. High bar for eligible underlyings
Only 29 underlyings were included in the first batch, with requirements of average daily options volume exceeding 150k contracts and market capitalization above $50 billion, limiting the scope of coverage.
2. Weak liquidity outside core hours
On the first day of extended trading, Tesla options saw total volume of just over 3,000 contracts during the extended session, with relatively high slippage costs.
3. Limited order types
Only limit orders are supported during extended hours; market and stop orders are not yet available, posing obstacles for adapting automated strategies.
III. Roadmap for boosting activity over the medium to long term
1. Continued expansion of eligible underlyings
The list of eligible underlyings will be updated every six months, with more highly liquid ETFs and leading stocks to be added gradually.
2. Supporting tools to launch alongside the expansion
Weekly options with morning expirations will launch on November 9, followed by binary options linked to the KPIs of publicly listed companies.
3. Full adoption by leading brokerages
Major platforms including TradeStation and Robinhood have already integrated extended-hours options trading, bringing retail users into the market. #WCTCS9

#特斯拉Q3财报 #SpaceX第三季度财报 SpaceX’s Q3 Results Blow Past Expectations! Musk Floats a $1 Trillion Revenue Target, Races Toward 13GW of Compute, While Tesla Merger Remains Uncertain!
After U.S. markets closed on October 20 Eastern Time, SpaceX (SPCXUS) delivered better-than-expected Q3 2026 results, with financial figures well above market forecasts and its stock rising nearly 8% at one point in after-hours trading.
This time, SpaceX’s growth engine is no longer just rockets and Starlink. Its Superintelligence (SI) business has surged, becoming a key driver of performance. On the earnings call, Musk outlined an aggressive plan: targeting around 13GW of AI compute by the end of 2027, with a chance of exceeding 20GW in an optimistic scenario, and envisioning an annual recurring revenue run rate (ARR) of $500 billion–$1 trillion by then. However, these are all long-term targets, not confirmed revenue guidance.
From losses to profits, AI business surges to become the main growth driver.
SpaceX’s Q3 total revenue was $14.575 billion, up 161% year over year and above market expectations of about $12.8 billion, beating estimates by more than 10%.
Profitability saw a significant turnaround: compared with a net loss of about $1.1 billion in the same period last year, the company posted net income of $4.4 billion this quarter; adjusted EBITDA was about $8.8 billion, up 306% year over year.
The biggest growth highlight came from the Superintelligence (SI) business: quarterly revenue of $7.56 billion, up about $5 billion sequentially and $6.7 billion year over year, making it SpaceX’s largest revenue segment. This included $7.022 billion in AI solutions and infrastructure revenue and $538 million in advertising revenue.
SI operating profit reached $431 million, a marked improvement from an operating loss of $1.257 billion in Q2. Company management said compute demand is exceptionally strong, and the payback period for newly deployed compute infrastructure at current prices has fallen to less than a year.
Musk used a “cheese shop” analogy to describe the current imbalance between compute supply and demand: hyperscalers and enterprises are continually seeking to rent compute, with strong demand for high-end chips such as GB300, GB200, and H100. “Every major hyperscaler and dozens of enterprises are chasing us for compute. Even if we said yes to all of them, we could rent out our compute several times over.”
Aggressive compute expansion, with 13GW targeted for 2027
The market had generally underestimated Musk’s ambitions in AI compute. The deployment timeline announced on this earnings call made a major impact.
GPU deployment pace: Currently, about 20,000 Nvidia GB300 chips come online each month, and management plans to increase monthly deployments to around 200,000, gradually transitioning to the Rubin architecture; it plans to complete a massive training cluster comprising 440,000 Vera Rubin computers by the end of Q2 2027 to support the Grok large language model.
Compute targets: The plan is to bring 2.5–3GW of compute online by the end of 2026. Musk said he is fairly confident about adding 10GW of compute in 2027, with total compute expected to reach around 13GW by year-end. More than 20GW is a low-probability optimistic scenario, not a firm target.
Infrastructure strategy: Musk said the bottlenecks to expanding AI superintelligence are not just chips, but also power, cooling, and supporting infrastructure. SpaceX is applying its aerospace engineering expertise to ground-based compute infrastructure in an effort to improve deployment efficiency.
In-house chip strategy: SpaceX and Tesla are collaborating on the Terafab chip manufacturing plan, targeting production of the first functional test chip in May–June 2027 and aiming for mass production in Q3 2028. The long-term vision is to build annual AI chip production capacity of 1 terawatt to reduce dependence on externally purchased chips.
The above timeline still faces considerable execution uncertainty.
Capital expenditure reached $19.593 billion in Q3, including $16.444 billion invested in the SI business. Management expects Q4 capex to rise further to $35 billion–$40 billion, continuing to support compute infrastructure construction.
Progress in satellite communications and Starship
Alongside ground-based AI compute, the traditional aerospace and Starlink businesses also made progress:
Starlink mobile business: Starlink Mobile surpassed 22 million monthly users. The company plans to expand U.S. mobile coverage through the acquisition of 800MHz spectrum resources. The deal still requires regulatory approval and completion, and whether it can become a mainstream mobile carrier remains to be seen. Enterprise and government businesses continue to grow and could become important sources of revenue in the future.
Satellite communications: Production of Starlink aviation terminals is expected to continue expanding before year-end; the additional downlink bandwidth from V3 satellites carried on a single Starship launch is about equivalent to 10 Falcon 9 Starlink launches. The long-term plan is to deploy about 100,000 high-bandwidth satellites to meet bandwidth demand from AI agents and significantly increase network capacity over the next three years.
Starship progress: The company plans to conduct its 15th flight next month, attempting to recover the booster and spacecraft simultaneously; it aims to gradually reach one launch per week in the first half of 2027, increasing to twice a week or even more frequently in the second half, advancing full rocket reusability and lowering space launch costs.
Tesla merger: A potential mega-merger, but far from a done deal.
The much-discussed potential merger between SpaceX and Tesla received a clear response on this earnings call: the two companies have established special committees to evaluate the possibility of a merger, but no transaction agreement has been reached. If ultimately completed, it could become one of the largest transactions in corporate merger history, with the exact size depending on the valuation and deal structure.
But Musk repeatedly highlighted the risks: due to his significant conflicts of interest, he is not participating in the special committee’s transaction discussions; there is no guarantee that a merger will happen. According to Musk’s remarks on the earnings call, Tesla would need support from holders of more than half of all outstanding shares, not just a majority of shares voted; the market view that “a merger is a done deal” is an overinterpretation, and considerable uncertainty remains.
Musk believes AI superintelligence is now entering a period of rapid development, and that merging the two companies could have long-term strategic value. Whether it ultimately goes ahead will still depend on negotiations between the two sides, shareholder approval, and relevant regulatory processes.$SPCX ‌

#GateWCTCS9全球交易赛 Spot gold: Spot gold surged and then retreated overnight. During yesterday’s US session, it briefly rebounded to around $4,155 before meeting resistance, then gave back its gains after midnight. The daily chart remains range-bound with a slight bearish bias.
During today’s Asian and European sessions, gold traded in a narrow $4,110–$4,125 range. The short-term rebound lacks momentum, and the overall bias remains range-bound to slightly bearish. Gold is currently caught between geopolitical support and interest-rate pressure. Middle East tensions, central bank gold purchases, and safe-haven demand provide support; however, high oil prices are lifting inflation expectations, markets are pricing in a relatively high chance of a December Fed rate hike, and elevated Treasury yields and the dollar continue to cap gold’s upside. Without a clear catalyst, gold is more likely to swing within the $4,065–$4,170 range.
Tonight, watch EIA crude oil inventories, remarks by Bank of England Governor Bailey and European Central Bank President Lagarde, as well as further market repricing of the interest-rate path ahead of the Fed’s upcoming policy meeting.
Tonight’s trading strategy: Aggressive short at 4145±5; conservative short at 4165±5. Target 4080; hold if it breaks! Aggressive long at 4102±5; conservative long at 4085±5. Target 4150; hold if it breaks!$XAUUSD ‌
For reference only; this is not investment advice!