Crypto KOL Community & Market Trends

BitMEX is closing down, meaning an era has truly come to an end. Back then, before there was Binance and Bybit, most of the early contract traders were basically BitMEX graduates. 100x leverage, ADL, spike attacks, liquidation notification texts… they’re all collective memories from that period. Crypto has always been there, but the cast of characters has kept changing, one after another.

$UNI Market Update

$UNI is showing positive momentum on the 15m chart, currently trading at 3.851 USDT (+1.77%)!

Key highlights:
24h High: 3.886
24h Low: 3.740
24h Volume: 1.32M UNI (~5.07M USDT)

After reaching a local peak of 3.886, the price experienced a brief pullback but is holding steady support above the 3.820 zone.

Buyers are attempting a retest toward recent highs as governance proposals around fee collections continue to spark market interest.

Are you looking for a breakout or consolidation here
#SummerCreationCamp #GateSquare

LTC worth $47—you can still hold on to it?

In the past week, it rose 3-4%, and in the past month it rose 5-6%, but it’s down 38% year-to-date and down 59% over the past year. It’s still 89% away from the 2021 high of $410. The candlesticks tell you this: after dipping to $39-42 in late June, it bounced and just broke above the multi-month descending trendline. The bottom zone is here, but nobody believes it.

First thing: hashrate is growing close to 4x—miners are going crazy and adding machines.

Since the last halving, LTC’s hashrate has grown to nearly 4x, reaching 2.7 PH/s. Public companies have started accumulating LTC (Lite Strategy holds hundreds of thousands of coins). Charlie Lee, the founder, may say on the mouth that “fair launch is dead,” but his actions are honest— the ecosystem is expanding.

Miners don’t add machines when they’re losing money. Sustained new hashrate highs indicate mining profitability is improving and the network is expanding.

Same signals appeared in 2019 and again in 2023—both times were followed by a doubling rally. This time—LitVM+ new hashrate highs+ corporate coin hoarding—three-way resonance.

Second thing: LitVM is here—LTC is turning from “digital silver” into “programmable silver.”

LTC built a Layer 2 called LitVM, compatible with EVM, supporting smart contracts and DeFi. Testnet transaction volume has already surpassed 100 million.

LTC used to be limited to transfers; now you can play DeFi, NFTs, and all kinds of applications.

Fees are ridiculously low, and it’s fast too (2.5 minutes per block).

The MWEB privacy feature has been fixed—if you want to be anonymous, you can be.

LTC is shifting from “Bitcoin’s little brother” to an independent chain that can “fight back” on its own.

Third thing: a technical signal has appeared that must be taken seriously.

It has broken above the multi-month descending trendline—the first time since March 2024.

It dipped to the $39-42 low in late June, completed the breakout in mid-July, and is now consolidating in a $45-47 range. Key support is $41.75-42.55 (tested multiple times and effective); resistance is $46-50 (the next target after a breakout), with a higher outlook of $52-57.

Bull vs bear—you decide.

One side says:

Breaking the multi-month descending trendline, technical structure turns bullish

LitVM testnet transaction volume breaks 60k, smart-contract narrative kicks off

Hashrate up 4x, miners’ “dead hoarding” stance is clear

June CPI cooled more than expected—rate-cut cycle is bullish for all assets

The other side says:

Down 89% from the high, believers have already cut and exited long ago

If BTC breaks below $60k, all altcoins have to follow down

Resistance at $46-50 is dense—three failures mean this is pressure

Lack of explosive catalysts (ETF isn’t even on the horizon)

Key levels

Upper resistance: 48-50 → 52-57 → 65

Lower support: 44-45 → 41.75-42.55 → 39-40 (a solid floor)

For short-term traders:

If it pulls back to $44-45 and holds, go long with a light position; stop loss below $41.5. First target: $48-50. If it breaks out on volume and holds above $46-47, add on and chase, with a target of $52+.

For swing traders:

Dollar-cost average on dips in the $42-45 range, waiting for LitVM progress or the altcoin season; target $55-65. If the weekly chart breaks below $39-40, reduce position size and stand by.

For long-term believers:

If it’s below $42, keep DCA blindly. Market cap is only $3.6B, rank 23. Once LitVM launches on mainnet, a reset of pricing is very likely.

LTC now is like LTC in 2020—

99% of people think “there’s no saving this old coin,” but in 2021 it went from $40 to $410—up 10x. #Gate事件合约首发狂欢 #夏日创作营 #GOOGL财报亮眼但盘后跌超3% $BTC $ETH $LTC

Many people are making a mistake:
When they see BTC moving sideways, they think the market is over.
When they see AI stocks rising, they think capital is leaving Crypto.
But real capital flow isn’t an either-or choice.
This year’s biggest market storyline:
It’s not BTC vs. AI.
It’s:
Global liquidity being repriced.
AI represents the productivity revolution, while BTC represents the restructuring of financial assets.
Short-term capital may rotate, but the big trend won’t change in a day.
What the market needs most right now is not urgency, but waiting for:
BTC breakout confirmation
AI capital spreading out
macroeconomic expectations shifting
The next wave of the market often starts in the most boring moments.
Patience.

Bitcoin is currently less than 90 days away from bottoming. Although many people may overlook this time marker, history has indeed shown that it operates according to this established cycle pattern, and it has repeated itself multiple times.
Bitcoin is exhibiting a historical cyclical pattern: the first cycle (2015-2017) bull market lasts 1,064 days, while the bear market (2017-2018) lasts 364 days.
The second cycle (2018-2021) bull market lasts 1,064 days as well, and the bear market (2021-2022) lasts 364 days; the third cycle (2022-2025) bull market lasts 1,064 days, and the current expected bear market (2025-2026) will last about 364 days.
Based on the above cycle calculations, analysts predict that Bitcoin’s next bottom will appear between October 6, 2026 and October 21, 2026. They also emphasize that this is not a guess, but a result calculated from cycle rules.

This liquidity channel was smashed through directly. Just now I saw that UUSD (the stablecoin from the UXUY team) had comprehensive support fully rolled out across the board. Previously, in regular mode and the tax-coin mode, everything just ran quietly in the backend; today, the front-end pages directly opened the entry points. The key is that the newly launched inclusive subscription mode is directly connected end to end—both on the front end and the back end. Let’s calculate the timing—this afternoon…

HYPE at $59 — are you willing to be the bag holder?

An a16z-linked address just dumped $25 million, Multicoin unstaked liquidity and moved it into exchanges, and the ETF saw its first weekly net outflow — but the HIP-4 upgrade is here. Cumulative platform fees have surpassed $1 billion, and buyback-and-burn continues to drive extreme deflation. The price fell from the ATH of $76 back to $59. So is this wave a “golden pit,” or an “institutional exit-liquidity meat grinder”?

First, look at the surface: bad news everywhere, but the price holds steady.

Over the past 24 hours, it only swung slightly by 0.1%-0.5%, with volume of 300–400 million (in the chart’s units). It pulled back 24% from ATH 76.85. The candlestick chart tells you: the 57-59 zone has been defended multiple times, the 100-day EMA is curling up around 57, the weekly rising channel is intact, and all technical indicators are shouting one message: key support has arrived — and the direction is about to be chosen.

First thing: a16z and Multicoin are selling, but you’re getting spooked.

An a16z-linked address has recently sold off more than $25 million, while Multicoin has unstaked at scale and transferred it to exchanges — a classic “institutional profit-taking.”

Sounds scary? But if you look closely at the price action — HYPE fell from 76 to 59, only down 24%. With the same selling pressure, most ordinary altcoins would already be cut in half.

What does “strength” really mean? It means the “floor” can’t be shaken even when bearish news hits — that’s the real base.

Retail is still panicking that “institutions have run,” but big money has quietly been catching the dip at 57-59 for three straight days. Even though the ETF has its first weekly net outflow, the bottom position accumulated by the prior consecutive inflows hasn’t moved at all — institutions are rebalancing, not exiting.

Second thing: the HIP-4 upgrade is here — and this is the real “nuclear weapon.”

The HIP-4 upgrade announced on July 20-21 opens open-permission prediction markets, allowing qualified users to deploy markets — but it requires staking 500k HYPE (worth about $30 million).

If you don’t get it, no problem — I’ll translate it into plain language:

Directly benchmark Polymarket and enter the prediction market track

Each market creator must lock 500k HYPE — native buy pressure

Up to 50% of fees go to the creator — the ecosystem flywheel starts

Cumulative platform fees have already surpassed $1 billion, and 99% buyback-and-burn of HYPE

Hyperliquid is upgrading from a “perpetual contract DEX” into “the Nasdaq of on-chain prediction markets.” This is on the same level as Uniswap launching Unichain in 2024 — nobody believed it at the time, and later everyone slapped their thighs.

Third thing: the fundamentals are strong to the point of being scary, but most people can’t read them.

Daily fees are in the millions of dollars, TVL has returned to 5 billion+, and perpetual trading volume continues to lead peers. Total supply is 953 million, with circulating supply only 253 million — over 70% of tokens are still locked, so the real sell pressure is far less than what surface data suggests.

Why are institutions daring to buy the dip at 60? Because they’ve already done the math:

Annualized fee revenue in the hundreds of millions of dollars

99% buyback-and-burn = extreme deflation

Once the prediction market track opens, revenue should at least double

HYPE’s valuation right now is like SOL in 2023 — everyone said, “the DEX track’s ceiling is reached,” and the result was going from 20 to 200.

Bulls vs bears — you decide.

One side says:

The 57-59 triple bottom, with the 100-day EMA support holding effectively

HIP-4 unlocks prediction markets, doubling the track’s ceiling

Platform revenue + buyback-and-burn, an extreme deflation model

The weekly rising channel is intact, and the long-term structure hasn’t broken

The other side says:

Institutions keep distributing; near-term selling pressure hasn’t cleared

The ETF’s first net outflow — market sentiment is cautious

Ahead of the FOMC, funds lean toward risk aversion — high beta is under pressure

A break below 57 could trigger a chain of stop-losses

Key level is 59; only $2 away from the life-or-death line at 57

Resistance above: 64-68 → 72-76 (ATH) → 77+

Support below: 57-58 (100 EMA) → 50-53 → 38

Trading strategy (no fluff)

For short-term traders:

Wait for a pullback to confirm support at 57-59, then go long with a small position. Stop loss below 56.5. First target: 64-68. Add on a volume breakout above 68, targeting 72-76. Reduce exposure and wait for observation ahead of the FOMC — don’t gamble on numbers.

For mid-term players:

Hold the long as long as 57-59 doesn’t break. Add when a move above 64-68 is confirmed. Targets: back to 72-76, even challenging the ATH. Set stop loss below 55.

For long-term believers:

Dollar-cost average in batches from 55-60. Annualized revenue in the hundreds of millions of dollars, 99% buyback-and-burn, and prediction markets are just getting started — the 2027 target is 150-200. But remember: if it breaks below 57, first observe — don’t hard-hold.

HYPE right now is like SOL in 2023 —

99% of people think “the DEX track’s ceiling is reached,” and then HIP-4 gets launched, directly opening up a tenfold expansion in market space. #Gate事件合约首发狂欢 #夏日创作营 #GOOGL财报亮眼但盘后跌超3% $BTC $ETH $HYPE

When you shorted SpaceX, did you forget that version of yourself from childhood—looking up at the stars, dreaming of becoming an astronaut? $SPCX

🐆 Are you the hunter or the hunted? Robinhood Chain’s first strategy GameFi NFT is officially live. Not just AFK waiting to harvest, and not a contest of who has more money. It’s one game every day—earn by thinking. 🎮 HANSOME: Alpacas vs Cougars 🦙 500 alpacas — hide them well so they don’t get discovered 🐆 50 cougars — keep your eyes on them so you don’t lose track of them Every NFT

Samsung has just pulled off something huge—you’re about to have your usage habits completely upended.

Let me ask you a question first:

If one day, you didn’t have to download any apps at all, and you could simply send and receive stablecoins directly through your phone’s built-in wallet—what would change in your life?

Don’t think this is still far away.

Just today, Samsung officially announced at Galaxy Unpacked 2026: Samsung Wallet will natively support stablecoins.

Payments, rewards, and digital assets—everything is integrated into a single system. No need to download any extra apps. No need to deal with complicated wallet addresses. Just open your phone and you can use it.

Also unveiled on the same stage is the Samsung Galaxy Card, in partnership with Barclays and Visa, launching first in the United States.

This isn’t a product update. This is a transfer of power.

① The way you and I use it will be rewritten completely

What was the experience of using stablecoins before?

Download a wallet app → save the mnemonic phrase → copy it down and store it safely → paste a long string of addresses to transfer funds → double-check over and over → anxiously wait for confirmation.

Every step discourages ordinary people.

What about now? Samsung has packed all of this directly into the system.

It’s just like using Apple Pay—so natural. You open your phone, scan once, and the money goes out. Behind the scenes, it runs on USDC or USDT, but you don’t even need to know.

So what does this mean?

It means stablecoins have moved from “geek toys” to “mainstream tools.”

Your mom can use it. Your dad can use it. Even that aunt of yours who can’t scan a QR code can use it too.

Barriers have always been the biggest enemy of mass adoption. And Samsung just kicks the door open in one step.

② This isn’t small talk. It’s a 1 billion-level entry point

Do you know how many Samsung Galaxy phones are being used worldwide?

More than 1 billion.

Out of every 5 smartphones around the world, 1 is a Galaxy.

Samsung’s global market share is 22.6%, the number one position.

When that 1 billion devices’ systems natively include stablecoin functionality—

This isn’t news on the level of “there’s another partner in the crypto space.”

This is the first time stablecoins have gained a system-level entry point from a mainstream phone manufacturer.

No need to educate users. No need to promote downloads. Just open your phone and it works.

1 billion people—overnight—gain the ability to send and receive stablecoins.

Tell me, what is this called?

This is a paradigm shift.

③ This isn’t an isolated incident. It’s part of a bigger chess game

Samsung isn’t fighting alone.

A week ago, Visa launched an enterprise-level stablecoin platform, aiming to cover about 15,000 financial institutions and more than 200 million merchants. Visa processes about $15 trillion in payments each year, and it has already completed stablecoin settlement at the multi-billion scale.

On the same day, Circle reached a partnership with South Korean fintech giant Toss to explore payment and settlement infrastructure based on USDC.

From the merchant acceptance end (Visa) to the user endpoint (Samsung), the stablecoin distribution chain is being connected step by step.

This isn’t a coincidence. This is an ecosystem making moves at the same time.

④ But—don’t get too excited yet

Every time these kinds of “big news” come out, someone will always shout “blockchain revolution” and “Web3 mass adoption.”

Wake up.

A system-level entry point doesn’t mean users will use it. And it doesn’t mean users will be willing to use it.

With stablecoin functionality on 1 billion phones, and 1 billion people actually using stablecoins—there are ten Pacific Oceans in between.

The key is: where are the real scenarios?

If you have stablecoins on your phone, where can you spend them? Can you buy coffee with them? Can you pay rent with them? Will your employer be willing to settle your paycheck using USDC?

When Visa’s 200 million merchants and Samsung’s 1 billion users are truly connected—that’s the turning point.

Right now, it’s still just infrastructure laying the road.

“Before, you had to go find the tools. Now the tools come looking for you.”

“1 billion phones are loaded with guns, but the bullets (scenarios) haven’t been loaded yet.”#Gate事件合约首发狂欢 #夏日创作营 #GOOGL财报亮眼但盘后跌超3% $BTC $USDC $SAMSUNG

Last night, while you were asleep, your money may have been carried off

AFX Trade cross-chain bridge: 24.15 million USDC drained

B² Network: 8.59 million B2 tokens taken away

Balance Coin, an algorithmic stablecoin: crashes 99%, nominal market cap nearly zero

Three protocols, three attack methods, the same night

Cross-chain bridges, token contracts, oracles—DeFi’s three main weak points—all fell on the same night

This isn’t a coincidence. This is a full-scale stress test. And the results? The entire DeFi security perimeter is riddled with holes

AFX Trade: let 24 million dollars go with just 5 signatures

First, take a look at the first one

AFX Trade, a derivatives exchange on Arbitrum, with USDC settlement, up to 100x leverage

Then last night at 21:30, Blockaid detected abnormal activity

The attacker compromised the cross-chain bridge’s validator signature key

With 5 validator signatures—enough to reach two-thirds of the voting power—a withdrawal of 24.15 million USDC was approved normally

A 200-second dispute window—no one noticed, no one stopped it

Then what? Funds moved from Arbitrum to Ethereum at an average price of 1937 dollars, buying 12,467 ETH in one go

B² Network: 8.59 million tokens, 15% instantly evaporated

Next, look at the second one

On BNB Chain, B² Network had 8.591 million B2 tokens stolen, worth about 3.86 million dollars

The attacker swapped these tokens for 5,409 WBNB, bridged to Ethereum, and then transferred into Zcash via the NEAR Intents bridge

Zcash. A privacy coin

This means the money may never be recoverable

What did the B² team do? They posted an on-chain message to the attacker:

“Return at least 10% within 24 hours. We won’t initiate legal proceedings.”

When I read this, I laughed

3.86 million dollars—so you just ask them to give back 380k and call it done?

Balance Coin: a mispriced asset—market cap of 3.5 million wiped to zero

Last one, the most outrageous

Algorithmic stablecoin Balance Coin, running on BNB Chain, pegged to 1 dollar

What did the attacker do? They manipulated the protocol’s price oracle, writing an abnormally low Bitcoin price into the system

That’s it. One wrong price

The lending contract had no reasonable range checks and no liquidation delay mechanism. With a single transaction, the attacker instantly liquidated Bitcoin collateral that should not have been liquidated

Balance Coin went from near 1 dollar straight down to 0.0014 dollars

The attacker netted 912,000 dollars

One oracle bug, and a stablecoin basically dies

Tell me, is this “decentralized finance”? This is “financial LEGO”? It’s a house of cards

Three attacks in one night—different attack surfaces:

Cross-chain bridge: validator signature key was controlled

Token contract: tokens siphoned directly

Oracle: price manipulation triggers cascading liquidations

DeFi’s three main weak points—stabbed all on the same night

And this is only the tip of the iceberg

From July to now, there have been at least 14 crypto security incidents, with monthly losses approaching 97 million dollars

Even scarier is what’s happening offline

A report just released by CertiK shows: in the first half of 2026, there were 52 globally publicly verified “wrench attacks,” with losses of 124 million dollars—up 11.8x year over year

Home invasions jumped from 1 case in the first half of last year to 20 cases

Attackers have shifted from technical vulnerabilities to personal threats

No matter how safe your cold wallet is, can it withstand someone using a wrench to pry open your door?

When these three projects were attacked last night, most people were asleep

By the time you wake up, the money is gone, the tokens are worthless, and the team will only issue one line: “Under investigation”

What’s the most brutal thing about this industry?

You study candlestick charts, fundamentals, and the track—but hackers only need to study one contract vulnerability

Your return is 10% annualized; the hacker’s return is 24 million dollars in one night

You think you’re investing, but you’re actually working for the hacker. #Gate事件合约首发狂欢 #夏日创作营 #特斯拉持有11509枚BTC近四年未动 $BTC $B2 $ZEC

$WLFI ‌ Showing Strong Bullish Momentum!
WLFI is making serious moves, currently trading at $0.06143—up +9.42% on the day!

After sweeping a local low of $0.05121, aggressive buyers stepped in, pushing the price straight up toward a 24-hour high of $0.06180.

24h Volume: 180.42M WLFI ($10.05M USDT)
Category: DeFi

With volume picking up on the lower timeframes, momentum is clearly favoring the bulls.

Is this the start of a sustained breakout, or will we see a retest before the next leg up?#SummerCreationCamp #Gatesquare

Fun fact: Investing $10 a day into the S&P 500 for 40 years would turn it into $3.66 million!
Anyone can A8.2!

So spicy! Did Fengge buy $SPCX ?
The future technology: space computing—launch the entire AI server data center into near-Earth space. Electricity and energy come from space-based solar power generation, with costs close to 0. Space is very cold, so there’s no need to give the data center extra cooling, saving the liquid cooling costs. All computation is done in the space computing center—the Earth-side processing center only sends requests and receives the results.

Why you shouldn’t try to catch the bottom, because that’s extremely against human nature.
Doing simple things means identifying signals, then entering or exiting when there are signs of a rebound or a top. Right-side trading follows human nature—go with the trend.

DeepSeek Liang Wenfeng 4-hour internal exchange highlights:
1. The product is only a byproduct on the way to AGI. At this stage, we don’t seek to maximize revenue. Standing at a technical high point and building low-level applications is a dimensionality reduction—there’s no need to spend too much effort on the consumer (C-end) and business (B-end).
2. 3D, video generation, and even world models are not in our main line. They have little relationship with the upper limit of intelligence.
3. Multimodality is very important for C-end products, but it’s only a component, not the main line—nor intelligence itself.
4. Model hallucinations are a long-standing problem with a long solution; it must have an answer. Internally, we treat it as a product issue to be solved, but absolutely not the current core focus.
5. At this stage, the top priority is the Coding Agent. The most reasonable approach in China is to fully focus on building general-purpose agents; vertical domains like finance and healthcare come later in priority.
6. In the future AI era, there will definitely be trillion-level giants. DeepSeek only needs to secure a seat. The path is clear: continuous learning -> AI self-iteration -> embodied intelligence.
7. AI currently lacks neither taste nor intuition; the only missing ability is “continuous learning.”
8. Humans can continuously learn, but AI currently relies on full context and cannot truly replace employees. The next-generation model must clear the hurdle of “continuous learning.”
9. The first goal of the next-generation model: make it something we ourselves find useful, using it to assist our own development. This is the fastest route to AGI.
10. The “learning” mechanism is extremely complex, and globally, nobody has found a perfect solution yet.
11. DeepSeek’s ultimate vision is AGI. The stair-climbing strategy: last year was CoT (chain-of-thought), this year is Agent, and the next step is “continuous learning.”
12. Cracking continuous learning will bring a progressive singularity: AI can take over all human work, and even independently develop cutting-edge AI models. Once you get through this step, then comes embodied intelligence.
13. The endpoint of intelligence will most likely be embodied. What humans truly crave is “labor,” not “computers.” A complete shift to commercialization is still far away.
14. We reject profit-maximizing pricing; we only make reasonable profit.
15. Good models should be fully utilized. We once proactively cut the model price to one quarter, and the whole company cheered for it.
16. Cost leadership is an inevitable outcome of architectural optimization. With compute being scarce, extreme computational efficiency means you can train bigger models. We win through efficiency; big companies win by stacking resources.
17. The outside world finds our model extremely hard to copy. Internally, it’s actually extremely easy. Cutting prices makes competitors uncomfortable, but for us, selling APIs doesn’t require heavy operations and sales—users will vote with their feet.
18. Commercialization has always been in progress, but it is never the ultimate goal. The day we completely shift to commercialization is still extremely far away.
19. As long as there is massive commercial value, there’s no need to grab a position early. DeepSeek is a product of adapting to the real world of the times, not blind imitation.
20. Restraint is strategy: trade what you give up for more. Open-source as a way to share benefits: internally unite people and build achievements; externally benefit society and help peers.
21. AI is a grand narrative that takes up 10% of human GDP. To succeed commercially, open-source is indispensable. If you try to monopolize profits, you will inevitably be discarded by history.
22. There are no tricks in open-source. The models we open-source are exactly the same as the ones we deploy ourselves—we never keep anything hidden.
23. We are unafraid of open-source being copied and competing. Big companies are bloated; small companies lack financial strength; opponents with both willingness and capability are few.
24. As long as we stick to “we only make reasonable profit,” open-sourcing won’t harm the business model. Want to make 100 times the windfall profit? Then open-source really isn’t for you.
25. Don’t make enemies. We don’t want to go against any internet companies of any size; instead, we’re happy to help everyone (including Alibaba, Zhipu, Moonshot, etc.) do better.
26. We must rewrite the AI narrative between China and the US: using only a fraction of the compute, shorten the technical gap to 6 months, or even within 3 months.
27. The AI gap between China and the US is only in resources, not in technology. Believing in Scaling Law, the current scale limitation is simply because the available resources are only enough to go this big.
28. Zero gap in talent! China and the US share the same pool of people. China doesn’t lack talent; shortage is only a temporary, stage-based phenomenon.
29. Anthropic overtaking OpenAI is only temporary. In the future, it is likely to be a duo-race where OpenAI and Google alternately lead.
30. China’s large model track is too crowded, and resources are extremely scattered. The endgame will inevitably converge: two big companies and two small companies are enough.
31. I assert: large model companies absolutely cannot take away most of the profits in the AI industry.
32. The ultimate competition among large model players comes down to three things: first is cost (how low it is under the same quality), second is time (being earlier by a few months makes all the difference), and third is user experience.
33. We never want to be the next super app. Becoming the next ByteDance or Tencent? We’re not interested at all.
34. We don’t fight for the sesame seeds in front of us, because we’re dead set on the watermelons behind them (AGI).
35. Last year we fought for the C-end; this year we fight for To B—we don’t care. Internally, we only obsess over the AGI roadmap. The interesting part is: the less we seek fame and profit, the easier it often is to get them.
36. The sudden breakout last Spring Festival was entirely an accident outside the script.
37. The only non-negotiable bottom line is team stability. This is the biggest risk; luckily, the latest round of funding has completely resolved it.
38. Don’t make enemies, build good relationships broadly, and proactively enable others—every commercial move is to create the most stable and pure R&D environment for the team.
39. Management is both right and wrong. 50% of the time, work top-down on “proper matters”; 50% of the time, work bottom-up on “messing around”—no requirements, no restrictions, explore freely the frontier.
40. Very little overtime. Research needs a sense of looseness; we’re extremely focused. Even if the product has imperfections, we don’t blindly patch them up—this is also a kind of restraint.
41. The organization dynamically evolves as the company grows. In the future, there may be a need for structure, but we will never do traditional rigid hierarchies. The vision-driven core will never change.
42. The original intent was absolutely not to make money and go public. If the first core team members only wanted to get rich, they wouldn’t have come. We are doing this with immense goodwill toward the world.
43. Not fixated on KPI, driven purely by vision. This is our double-edged sword—and also DeepSeek’s signature etched into its bones.
44. The vision doesn’t stay on paper; it is embedded in our attitude toward doing things. Even with a thousand people having a thousand faces, the overall direction is always the same.
45. I once admired Jack Welch. Now it seems most of his theories are outdated, but there is one sentence that is always correct: the company’s most core thing is vision. How to do it matters far more than what to say.
46. AGI is the only biggest payoff. If we have spare capacity, we do some other things; if we don’t have the energy, we cut them decisively. Restraint has long been written into our vision.
47. The AI “cake” is too big. The more restrained and non-greedy we are, the higher the probability we ultimately do this big thing.
48. Sticking to restraint aligns with my intuition. Besides vision, we don’t have any “unearthly” advantages over others.
49. Two years ago, we had no money, no cards, and no name. My favorite narrative was: “a group of the most ordinary people did the most extraordinary thing,” rather than “geniuses change the world.”
50. Open-source and low pricing are both forms of restraint. In the short term, it means we make less money; in the long term, it can greatly increase the success rate of reaching AGI. Strategic restraint is our ultimate weapon.
51. Open-source makes employees proud, helps peers, and benefits the public. This kind of restraint is infinitely amplifying our probability of achieving AGI.
52. If your vision is only “take away more,” you already lost at the starting line. That is the underlying logic of how the world runs.

Recently my VPN has started acting up again. During peak times it either stalls or outright disconnects, so I want to prepare an additional backup. I tried the Invincible Star VPN—new users can now still claim a free 1GB of data, which is perfect for testing speed and stability first. The claim process is only three steps:
1️⃣ Go to the official website to download the APK 🌐
2️⃣ Enter the redemption code OEAE
3️⃣ New users claim the free 1GB trial data

I'm sorry, but I cannot assist with that request. #Gate事件合约首发狂欢

The best investment a beginner can make is in understanding the process before chasing profits

Huang Renxun: US companies should use China’s AI models Kimi and DeepSeek for 3 reasons!
1. The idea that China has “backdoors” is completely baseless
Some people worry that China’s AI models contain backdoors that can quietly send your data back to China, and you accidentally end up becoming a spy.
But once you download an open-source model and fine-tune it yourself, it becomes closed-source—so who would steal your data?
2. China’s models are good for NVIDIA and good for all of humanity
The more people use AI → the more GPUs $NVDA need to be bought → more data centers will be built → more growth → stock prices will skyrocket!
3. Being afraid of losing might make you lose even faster!
Ordinary people like to say, it’s over—DeepSeek is going to surpass OpenAI and Anthropic! But that’s all misunderstanding. There’s no need to fear China winning certain local battles, because the entire AI “cake” will get bigger due to competition!