正在加载视频...

视频加载失败

返回首页

Core steps its block reward down every year instead of halving it Core DAO 🔶 caps coredaoorg:native at 2.1 billion tokens, a figure its documentation describes as exactly 100 times Bitcoin's 21 million. New supply enters through block rewards, the tokens paid to the validators producing each block, spread...

BSCNews's profile picture

BSCN

1,371,983 subscribers

106,065 次观看 • 11 天前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Laguna S 2.1 performs at GLM-5.2 level on building popular games with 6x fewer params! We gave three local models the same task: build three popular arcade games that play themselves. Each game is one self-contained HTML file with a bot that plays it. Prompts: – Geometry Dash – Doodle Jump – Air Hockey Outputs: Laguna S 2.1: 10.3K tokens GLM-5.2: 26.4K tokens Hy3: 10.4K tokens Laguna held its quality against a 753B model. We think Laguna's Geometry Dash looked the best of the three, the cube clears every spike and block. GLM won Air Hockey. Its table looked the most detailed of all. Hy3 was the only model that added shooting to its Doodle Jump. But Laguna is the only model in our benchmark that runs on a MacBook with 128GB!
0:19

Laguna S 2.1 performs at GLM-5.2 level on building popular games with 6x fewer params! We gave three local models the same task: build three popular arcade games that play themselves. Each game is one self-contained HTML file with a bot that plays it. Prompts: – Geometry Dash – Doodle Jump – Air Hockey Outputs: Laguna S 2.1: 10.3K tokens GLM-5.2: 26.4K tokens Hy3: 10.4K tokens Laguna held its quality against a 753B model. We think Laguna's Geometry Dash looked the best of the three, the cube clears every spike and block. GLM won Air Hockey. Its table looked the most detailed of all. Hy3 was the only model that added shooting to its Doodle Jump. But Laguna is the only model in our benchmark that runs on a MacBook with 128GB!

atomic.chat

66,917 次观看 • 1 个月前

⚡️BREAKDOWN: Bondholders are offering the US Treasury about 10x its long-end buyback cap. Treasury runs liquidity support buybacks in the 10 to 30 year sectors, capped at $2 BILLION per operation. On Tuesday holders offered $19.87 BILLION. Treasury ended up taking $2 BILLION. Across 22 long-end operations this year, holders have offered more than $520 BILLION against a combined cap of $44 BILLION. When the first long-end operation ran in June 2024, offers covered the cap 1.93 times. On Tuesday they covered it 9.93 times. So Treasury doubled the cap, to at least $4 BILLION per operation from September 9 through November 4. Its stated reason was the "significant volume of high-quality offers" it keeps receiving. Even at $4 BILLION, that is negligible in a market worth $32.2 TRILLION, per Reuters. For context, Treasury launched the 2000 buyback program during an era of budget surpluses. That fiscal year closed with a $237 BILLION surplus, the largest ever recorded in nominal terms, as the government paid down publicly held debt. Today the stated purpose is liquidity support, the deficit is projected at around $2 TRILLION, and the national debt has crossed $40 TRILLION.
0:14

⚡️BREAKDOWN: Bondholders are offering the US Treasury about 10x its long-end buyback cap. Treasury runs liquidity support buybacks in the 10 to 30 year sectors, capped at $2 BILLION per operation. On Tuesday holders offered $19.87 BILLION. Treasury ended up taking $2 BILLION. Across 22 long-end operations this year, holders have offered more than $520 BILLION against a combined cap of $44 BILLION. When the first long-end operation ran in June 2024, offers covered the cap 1.93 times. On Tuesday they covered it 9.93 times. So Treasury doubled the cap, to at least $4 BILLION per operation from September 9 through November 4. Its stated reason was the "significant volume of high-quality offers" it keeps receiving. Even at $4 BILLION, that is negligible in a market worth $32.2 TRILLION, per Reuters. For context, Treasury launched the 2000 buyback program during an era of budget surpluses. That fiscal year closed with a $237 BILLION surplus, the largest ever recorded in nominal terms, as the government paid down publicly held debt. Today the stated purpose is liquidity support, the deficit is projected at around $2 TRILLION, and the national debt has crossed $40 TRILLION.

Coin Bureau

36,987 次观看 • 18 天前

🚨 JUST IN: WORLD FOUNDATION SELLS $65 MILLION IN WLD TOKENS THROUGH OTC DEALS AS TOKEN HITS ALL-TIME LOW Sam Altman's World Foundation disclosed that its subsidiary World Assets sold roughly 239 million WLD tokens at an average price of $0.2719 per token through OTC transactions with four counterparties over the past week. Of the $65 million raised, $25 million worth of tokens are locked for six months to prevent immediate selling by buyers. Proceeds are earmarked for core operations, R&D, Orb manufacturing, and ecosystem expansion. WLD is currently trading near its all-time low of $0.27, down over 97% from its March 2024 peak of $11.74.
0:15

🚨 JUST IN: WORLD FOUNDATION SELLS $65 MILLION IN WLD TOKENS THROUGH OTC DEALS AS TOKEN HITS ALL-TIME LOW Sam Altman's World Foundation disclosed that its subsidiary World Assets sold roughly 239 million WLD tokens at an average price of $0.2719 per token through OTC transactions with four counterparties over the past week. Of the $65 million raised, $25 million worth of tokens are locked for six months to prevent immediate selling by buyers. Proceeds are earmarked for core operations, R&D, Orb manufacturing, and ecosystem expansion. WLD is currently trading near its all-time low of $0.27, down over 97% from its March 2024 peak of $11.74.

BSCN

66,269 次观看 • 5 个月前

Sold 32 coins. Bought 1,550. 48 times more, at a 15% discount, into the crash the market blamed on the sale. Strategy disclosed today that while everyone panicked over its $2.5 million Bitcoin sale, it was quietly buying the dip that panic created. 1,550 Bitcoin for $101 million, at $65,332 a coin, far below the $77,135 it sold for and below its own cost basis. The bears called the sale the first crack, a forced liquidation, the start of the death spiral. The answer was a buy 48 times the size of the sale that scared them. This is the machine we described: a state-contingent allocator. Above its funding line, it turns market access into Bitcoin. The sale was the exception. The buy is the rule. It also closed the question the sale opened. The cash reserve behind the preferred dividends had thinned to $900 million, about six months of cover. He rebuilt it to $1 billion in the same week. But watch how, because that is the real story. He funded none of it with coins. He funded it with $181 million of freshly issued stock, then spent it on Bitcoin and the reserve. The coins were never the funding source. The equity is. That is the flywheel working exactly as built, and the cost of it surfacing at the same time. Every turn now runs on issuing shares, and the premium that once made each share buy more Bitcoin than it diluted has compressed hard. He bought low. He sold his own stock low to do it. So the question quietly turns. It was never whether Saylor sells his Bitcoin. He just proved again that he buys far more than he sells. It is what each turn of the engine now costs in dilution, and how long the market keeps paying a premium worth that cost. He bought the dip. The dip was partly his own making. And he paid for it in equity, not coins.
0:10

Sold 32 coins. Bought 1,550. 48 times more, at a 15% discount, into the crash the market blamed on the sale. Strategy disclosed today that while everyone panicked over its $2.5 million Bitcoin sale, it was quietly buying the dip that panic created. 1,550 Bitcoin for $101 million, at $65,332 a coin, far below the $77,135 it sold for and below its own cost basis. The bears called the sale the first crack, a forced liquidation, the start of the death spiral. The answer was a buy 48 times the size of the sale that scared them. This is the machine we described: a state-contingent allocator. Above its funding line, it turns market access into Bitcoin. The sale was the exception. The buy is the rule. It also closed the question the sale opened. The cash reserve behind the preferred dividends had thinned to $900 million, about six months of cover. He rebuilt it to $1 billion in the same week. But watch how, because that is the real story. He funded none of it with coins. He funded it with $181 million of freshly issued stock, then spent it on Bitcoin and the reserve. The coins were never the funding source. The equity is. That is the flywheel working exactly as built, and the cost of it surfacing at the same time. Every turn now runs on issuing shares, and the premium that once made each share buy more Bitcoin than it diluted has compressed hard. He bought low. He sold his own stock low to do it. So the question quietly turns. It was never whether Saylor sells his Bitcoin. He just proved again that he buys far more than he sells. It is what each turn of the engine now costs in dilution, and how long the market keeps paying a premium worth that cost. He bought the dip. The dip was partly his own making. And he paid for it in equity, not coins.

Shanaka Anslem Perera ⚡

142,587 次观看 • 3 个月前

Did you know the James Webb Space Telescope isn’t orbiting Earth at all? The James Webb Space Telescope operates near the Sun–Earth L2 point, about 1.5 million km from Earth. From there, the Sun, Earth, and Moon stay on the same side of the telescope, allowing its giant sunshield to block their heat and light at the same time. That keeps Webb cold and stable enough to detect extremely faint infrared signals from deep space. Because of this, Webb can peer through clouds of cosmic dust and observe galaxies whose light has traveled for more than 13 billion years. Every image it captures is not just a view across space, but a look back into the early history of the universe.
0:12

Did you know the James Webb Space Telescope isn’t orbiting Earth at all? The James Webb Space Telescope operates near the Sun–Earth L2 point, about 1.5 million km from Earth. From there, the Sun, Earth, and Moon stay on the same side of the telescope, allowing its giant sunshield to block their heat and light at the same time. That keeps Webb cold and stable enough to detect extremely faint infrared signals from deep space. Because of this, Webb can peer through clouds of cosmic dust and observe galaxies whose light has traveled for more than 13 billion years. Every image it captures is not just a view across space, but a look back into the early history of the universe.

Cosmos Archive

22,557 次观看 • 3 个月前

The biggest Bitcoin miners on earth are quietly walking away from mining Bitcoin, and the reason is not the one everyone keeps repeating. They are not fleeing a dead business. They lost an auction for their own power, and the winner was artificial intelligence. Start with the brutal arithmetic. It now costs the average public miner around $80,000 in cash to produce a single Bitcoin, and for stretches of this year $BTC traded below that. The most efficient operators on the cheapest power still clear a margin, but an estimated 15 to 20 percent of the global fleet is mining at a loss right now, burning more in power than the coins are worth the second they are minted. Three straight downward difficulty adjustments earlier this year, the first such streak since 2022, were the footprint of machines going dark. That looks like a simple story of a broken business until you see the number that explains the exodus. The same megawatt of power that earns a Bitcoin miner roughly $1 million a year earns between $10 and $20 million a year hosting AI compute. Ten to twenty times more, for the identical electricity, substation, and cooling. What made industrial miners valuable was never the mining. It was the power contracts, the land, the grid interconnects. AI walked in and bid an order of magnitude higher for exactly those assets. Mining did not fail. It got outbid for its own infrastructure. When Core Scientific runs its BTC segment at a negative margin while its AI colocation business prints money, the decision writes itself. CoinShares estimates listed miners could pull up to 70 percent of their revenue from AI by year end, up from about 30 percent. The power is being repriced to its highest use, and Bitcoin lost the bidding. If the giants leave, what happens to the network they secured? The doom posts assume it weakens. It does not, because Bitcoin has a self-healing reflex written into its core. When miners switch off, blocks slow, and within two weeks difficulty automatically drops, which makes mining cheaper and more profitable for everyone still running. The security does not vanish, it relocates, and you can already see where. State-backed pools are appearing, with one Gulf operator reportedly standing up a national pool near 3 percent of global hashrate, alongside private fleets and the handful of public miners like Marathon still choosing to buy Bitcoin rather than lease their power away. The network even hit an all-time high above one zettahash this year as the pivot accelerated. It does not need any particular miner. It needs someone, somewhere, for whom the math still works, and cheap stranded power has no shortage of those. But there is a deeper timer here, and the AI pivot just exposed it. Today miners earn almost everything from the block subsidy and almost nothing from fees, often under one percent of revenue on a quiet day. That subsidy halves again in 2028, and every four years after, marching toward zero. For Bitcoin to pay for its own security forever, fees eventually have to replace it. The open question is whether they can, and the evidence cuts both ways. On busy days, during token launches and inscription waves, fees have already spiked past 15 percent of revenue, and in 2024 some blocks earned more in fees than the entire subsidy. The capacity is there in bursts. Whether bursts become a baseline is the single most important unanswered question in Bitcoin. The AI exodus did not create that question. It pulled the cover off it years early, and showed how fast capital abandons hashing the moment something pays more. So the honest read is not that AI kills Bitcoin mining. It is stranger than that. AI is the first bidder rich enough to reveal what Bitcoin's security was always quietly worth, and what it will cost to keep once the free coins stop coming. The miners are not abandoning a sinking ship. They are selling the deck to a higher bidder while the same clock everyone forgot about keeps ticking underneath.
0:10

The biggest Bitcoin miners on earth are quietly walking away from mining Bitcoin, and the reason is not the one everyone keeps repeating. They are not fleeing a dead business. They lost an auction for their own power, and the winner was artificial intelligence. Start with the brutal arithmetic. It now costs the average public miner around $80,000 in cash to produce a single Bitcoin, and for stretches of this year $BTC traded below that. The most efficient operators on the cheapest power still clear a margin, but an estimated 15 to 20 percent of the global fleet is mining at a loss right now, burning more in power than the coins are worth the second they are minted. Three straight downward difficulty adjustments earlier this year, the first such streak since 2022, were the footprint of machines going dark. That looks like a simple story of a broken business until you see the number that explains the exodus. The same megawatt of power that earns a Bitcoin miner roughly $1 million a year earns between $10 and $20 million a year hosting AI compute. Ten to twenty times more, for the identical electricity, substation, and cooling. What made industrial miners valuable was never the mining. It was the power contracts, the land, the grid interconnects. AI walked in and bid an order of magnitude higher for exactly those assets. Mining did not fail. It got outbid for its own infrastructure. When Core Scientific runs its BTC segment at a negative margin while its AI colocation business prints money, the decision writes itself. CoinShares estimates listed miners could pull up to 70 percent of their revenue from AI by year end, up from about 30 percent. The power is being repriced to its highest use, and Bitcoin lost the bidding. If the giants leave, what happens to the network they secured? The doom posts assume it weakens. It does not, because Bitcoin has a self-healing reflex written into its core. When miners switch off, blocks slow, and within two weeks difficulty automatically drops, which makes mining cheaper and more profitable for everyone still running. The security does not vanish, it relocates, and you can already see where. State-backed pools are appearing, with one Gulf operator reportedly standing up a national pool near 3 percent of global hashrate, alongside private fleets and the handful of public miners like Marathon still choosing to buy Bitcoin rather than lease their power away. The network even hit an all-time high above one zettahash this year as the pivot accelerated. It does not need any particular miner. It needs someone, somewhere, for whom the math still works, and cheap stranded power has no shortage of those. But there is a deeper timer here, and the AI pivot just exposed it. Today miners earn almost everything from the block subsidy and almost nothing from fees, often under one percent of revenue on a quiet day. That subsidy halves again in 2028, and every four years after, marching toward zero. For Bitcoin to pay for its own security forever, fees eventually have to replace it. The open question is whether they can, and the evidence cuts both ways. On busy days, during token launches and inscription waves, fees have already spiked past 15 percent of revenue, and in 2024 some blocks earned more in fees than the entire subsidy. The capacity is there in bursts. Whether bursts become a baseline is the single most important unanswered question in Bitcoin. The AI exodus did not create that question. It pulled the cover off it years early, and showed how fast capital abandons hashing the moment something pays more. So the honest read is not that AI kills Bitcoin mining. It is stranger than that. AI is the first bidder rich enough to reveal what Bitcoin's security was always quietly worth, and what it will cost to keep once the free coins stop coming. The miners are not abandoning a sinking ship. They are selling the deck to a higher bidder while the same clock everyone forgot about keeps ticking underneath.

Shanaka Anslem Perera ⚡

90,771 次观看 • 2 个月前

ReLU vs Leaky ReLU 👉 = ReLU = ReLU is the default activation in modern deep learning — cheap to compute, and stable enough to train networks hundreds of layers deep. To see what it does, picture five boba tea shops on the same block — 𝚊, 𝚋, 𝚌, 𝚍, 𝚎 — each running their own books. Each value is a shop's monthly profit — receipts minus rent, ingredients, and wages. When profit is positive, the shop stays open and the owner pockets every dollar. When profit turns negative, the shop runs out of cash and shutters — the lights go off, the books are wiped to zero. ReLU is exactly that rule, applied one shop at a time. Read the diagram left to right. The first column is the raw value x — each shop's profit at month's end. The second column is the gate: 1 if the shop is open (x > 0), 0 if it has shuttered. The last column is the ReLU output: open shops pass their profit through untouched, while shuttered ones are zeroed out. Five rows means five parallel shops on the same block, each evaluated independently. That's why ReLU is called an element-wise activation: every neuron decides its own fate. = LeakyRelu = Plain ReLU wipes negative values to zero — clean, but a shop that shutters can never recover, since both its output and its gradient stay pinned at zero. This is the dying ReLU problem, and in deep networks it can quietly kill a meaningful fraction of the units. Leaky ReLU is the one-line fix: instead of shuttering, the shop files for Chapter 11 protection and keeps the lights on at reduced capacity. Its debt is restructured down to a fraction α (typically 0.1) — the rest is forgiven, and the shop is wounded, not killed. A small negative signal still flows through, so the gradient survives, and the shop can crawl back to life if a TikTok goes viral. Read the diagram left to right. The first column is the raw value x — each shop's profit at month's end. The second column is the leakage α — the fraction of the loss held over after restructuring (default 0.1, editable). The third column is the gate: 1 for shops still in the black, α for those operating under bankruptcy protection. The last column is the Leaky ReLU output: y = x · gate. Profitable shops pass through untouched; struggling ones shrink by a factor of α but still carry a sign. Five rows means five parallel shops, each evaluated independently. Like ReLU, this is an element-wise activation: every neuron's fate is decided on its own merits. #aibyhahd
0:18

ReLU vs Leaky ReLU 👉 = ReLU = ReLU is the default activation in modern deep learning — cheap to compute, and stable enough to train networks hundreds of layers deep. To see what it does, picture five boba tea shops on the same block — 𝚊, 𝚋, 𝚌, 𝚍, 𝚎 — each running their own books. Each value is a shop's monthly profit — receipts minus rent, ingredients, and wages. When profit is positive, the shop stays open and the owner pockets every dollar. When profit turns negative, the shop runs out of cash and shutters — the lights go off, the books are wiped to zero. ReLU is exactly that rule, applied one shop at a time. Read the diagram left to right. The first column is the raw value x — each shop's profit at month's end. The second column is the gate: 1 if the shop is open (x > 0), 0 if it has shuttered. The last column is the ReLU output: open shops pass their profit through untouched, while shuttered ones are zeroed out. Five rows means five parallel shops on the same block, each evaluated independently. That's why ReLU is called an element-wise activation: every neuron decides its own fate. = LeakyRelu = Plain ReLU wipes negative values to zero — clean, but a shop that shutters can never recover, since both its output and its gradient stay pinned at zero. This is the dying ReLU problem, and in deep networks it can quietly kill a meaningful fraction of the units. Leaky ReLU is the one-line fix: instead of shuttering, the shop files for Chapter 11 protection and keeps the lights on at reduced capacity. Its debt is restructured down to a fraction α (typically 0.1) — the rest is forgiven, and the shop is wounded, not killed. A small negative signal still flows through, so the gradient survives, and the shop can crawl back to life if a TikTok goes viral. Read the diagram left to right. The first column is the raw value x — each shop's profit at month's end. The second column is the leakage α — the fraction of the loss held over after restructuring (default 0.1, editable). The third column is the gate: 1 for shops still in the black, α for those operating under bankruptcy protection. The last column is the Leaky ReLU output: y = x · gate. Profitable shops pass through untouched; struggling ones shrink by a factor of α but still carry a sign. Five rows means five parallel shops, each evaluated independently. Like ReLU, this is an element-wise activation: every neuron's fate is decided on its own merits. #aibyhahd

Tom Yeh

32,561 次观看 • 4 个月前

32 coins. $2.5 million. 0.0038% of the stack. That is the sale the market is now blaming for a $3 billion liquidation cascade and a Bitcoin price nearly halved from its peak. A $2.5 million sale cannot move a trillion-dollar asset. It is a rounding error. In the same week, Strategy raised $128.3 million selling its own stock, 50 times larger. It did not need to sell coins. It chose to. The crash has real drivers: a record 13-day run of ETF outflows, a rotation into AI, a Fed in no hurry to cut. But the accelerant the market keeps naming is 32 coins. The coins were never the point. The signal was. And the signal was deliberate. Michael Saylor told the Q1 call he would “probably sell some bitcoin to pay a dividend just to inoculate the market and send the message that we did it.” His logic was sound: prove the Bitcoin is usable capital, not a vault that can never be opened, and show he is not a prisoner of his own vow. His “never sell” always meant be a net accumulator. He is up more than 170,000 coins this year against the 32 he sold, and he scores himself on one number, Bitcoin per share. By that math, defending the dividend with a sliver was discipline, not distress. The market read it as the opposite. The dose became the catalyst now blamed for the crash. The inoculation became the infection. Because what changed was never Strategy’s solvency. It was its identity. The market has stopped pricing a permanent holder and started pricing what the filings always described: a state-contingent allocator now funding its own preferred dividends, at the margin, from the Bitcoin beneath them. And the buffer is thinning. The cash reserve behind those dividends has fallen from $2.25 billion to $900 million. Against a preferred bill near $1.7 billion a year, that is roughly 6 months of runway. Be precise. This is not a death spiral. Strategy still holds 843,706 Bitcoin, worth more than $50 billion even now, and has more funding levers than almost any company alive. A real rally makes this a footnote, and the sell-side calling the reaction overdone is not wrong on the fundamentals. But the regime has changed. The question is no longer Bitcoin’s price on any given day. It is the cadence of the dividend declarations and the path of that reserve. Bitcoin did not acquire a yield. The wrapper acquired liabilities. This week the market learned that difference costs far more than 32 coins.
0:10

32 coins. $2.5 million. 0.0038% of the stack. That is the sale the market is now blaming for a $3 billion liquidation cascade and a Bitcoin price nearly halved from its peak. A $2.5 million sale cannot move a trillion-dollar asset. It is a rounding error. In the same week, Strategy raised $128.3 million selling its own stock, 50 times larger. It did not need to sell coins. It chose to. The crash has real drivers: a record 13-day run of ETF outflows, a rotation into AI, a Fed in no hurry to cut. But the accelerant the market keeps naming is 32 coins. The coins were never the point. The signal was. And the signal was deliberate. Michael Saylor told the Q1 call he would “probably sell some bitcoin to pay a dividend just to inoculate the market and send the message that we did it.” His logic was sound: prove the Bitcoin is usable capital, not a vault that can never be opened, and show he is not a prisoner of his own vow. His “never sell” always meant be a net accumulator. He is up more than 170,000 coins this year against the 32 he sold, and he scores himself on one number, Bitcoin per share. By that math, defending the dividend with a sliver was discipline, not distress. The market read it as the opposite. The dose became the catalyst now blamed for the crash. The inoculation became the infection. Because what changed was never Strategy’s solvency. It was its identity. The market has stopped pricing a permanent holder and started pricing what the filings always described: a state-contingent allocator now funding its own preferred dividends, at the margin, from the Bitcoin beneath them. And the buffer is thinning. The cash reserve behind those dividends has fallen from $2.25 billion to $900 million. Against a preferred bill near $1.7 billion a year, that is roughly 6 months of runway. Be precise. This is not a death spiral. Strategy still holds 843,706 Bitcoin, worth more than $50 billion even now, and has more funding levers than almost any company alive. A real rally makes this a footnote, and the sell-side calling the reaction overdone is not wrong on the fundamentals. But the regime has changed. The question is no longer Bitcoin’s price on any given day. It is the cadence of the dividend declarations and the path of that reserve. Bitcoin did not acquire a yield. The wrapper acquired liabilities. This week the market learned that difference costs far more than 32 coins.

Shanaka Anslem Perera ⚡

165,572 次观看 • 3 个月前

RealFi’s Global Payment Rewards Launch Arrives in 48 Hours Only Two days remain until RealFi officially launches its Payment Rewards platform globally on January 9, 2026, expanding access to both retail users and institutions worldwide. RealFi’s Payment Rewards Solution enables users and institutions to submit receipts from purchases made with Bitcoin, Ethereum, and XRP, earning rewards or cashback in Real Tokens. Learn more about RealFi and Real Token at The Real Token is a core component of the RealFi ecosystem, functioning as the cashback and rewards mechanism across all supported payment transactions.
0:10

RealFi’s Global Payment Rewards Launch Arrives in 48 Hours Only Two days remain until RealFi officially launches its Payment Rewards platform globally on January 9, 2026, expanding access to both retail users and institutions worldwide. RealFi’s Payment Rewards Solution enables users and institutions to submit receipts from purchases made with Bitcoin, Ethereum, and XRP, earning rewards or cashback in Real Tokens. Learn more about RealFi and Real Token at The Real Token is a core component of the RealFi ecosystem, functioning as the cashback and rewards mechanism across all supported payment transactions.

RealFI

398,005 次观看 • 8 个月前

A new 'Jesus-centric' mobile network is launching in the US and it plans to block LGBTQ+ content. Radiant Mobile says its service will filter more than 100 categories of online material, including p***, gaming and violence. But it will also give users the option to block content related to LGBTQ+ and trans issues. Founder Paul Fisher said the goal is to create an online space, saying in an interview with MIT Technology Review, “we are going to create—and we think we have every right to do so—an environment that is Jesus-centric... [that is] void of LGBT, void of trans.” The network operates as a mobile virtual network operator, using infrastructure from T-Mobile, and has reportedly partnered with churches and Christian influencers to promote the service.
0:16

A new 'Jesus-centric' mobile network is launching in the US and it plans to block LGBTQ+ content. Radiant Mobile says its service will filter more than 100 categories of online material, including p***, gaming and violence. But it will also give users the option to block content related to LGBTQ+ and trans issues. Founder Paul Fisher said the goal is to create an online space, saying in an interview with MIT Technology Review, “we are going to create—and we think we have every right to do so—an environment that is Jesus-centric... [that is] void of LGBT, void of trans.” The network operates as a mobile virtual network operator, using infrastructure from T-Mobile, and has reportedly partnered with churches and Christian influencers to promote the service.

PinkNews

12,014 次观看 • 4 个月前

Category Labs is proud to introduce Cadence, our multiple-concurrent-proposers (MCP) consensus protocol that matches the optimal good-case latency of single-leader consensus while supporting arbitrarily short block intervals. When combined with BTX, our design for encrypted mempools, this represents a significant step towards solving the problem of MEV at the protocol level. In nearly every blockchain today, a single party ends up in control of each block: it decides which transactions get in, and can reorder them at will. MCP is the natural fix, but most recent designs pay for it with a separate aggregation phase, adding two extra communication rounds per block. Cadence makes the proposers part of consensus itself. Its fast path finalizes in an optimal three communication rounds, even when proposers are offline. Cadence also offers speculative finality, similar to MonadBFT, after just two rounds, revertible only if a proposer provably equivocated. In a simulation using estimated network delays between Monad mainnet's 200 globally distributed validators, finalization takes 219 ms on average, speculative finality 167 ms. Cadence pushes pipelining to the extreme: each block is proposed and finalized in its own independent consensus instance, without waiting on preceding blocks. The block interval then becomes a protocol parameter that can be arbitrarily small. At our initial target of 100 ms, a transaction waits on average just 50 ms to enter a proposal, and oracle prices, liquidations, and auctions can update every 100 ms. Cadence dynamically throttles the opening of new instances to bound the number of outstanding slots even during periods of network instability. When the network is healthy (under synchrony), a transaction included by an honest proposer can be neither dropped nor deferred (short-term censorship resistance), and no proposer can see the others' proposals in time to react (hiding). We prove both, together with safety and liveness under partial synchrony at the optimal 3f+1 fault bound. The Cadence protocol is modular: each module is simple on its own, and any of them can be swapped out without touching the rest. Cadence also builds on components already being deployed: proposals are disseminated as erasure-coded chunks over Deterministic RaptorCast, now rolling out on Monad, and validators vote on proposal digests, so voting does not wait for the full data to arrive. Start with the interactive tutorial: Full paper: Joint work by Kushal Babel, Fatima Elsheimy, Lioba Heimbach, Mohammad Mussadiq Jalalzai, Tobias Klenze, Jovan Komatovic, Jason Milionis, Mike Setrin, and Victor Shoup.
0:13

Category Labs is proud to introduce Cadence, our multiple-concurrent-proposers (MCP) consensus protocol that matches the optimal good-case latency of single-leader consensus while supporting arbitrarily short block intervals. When combined with BTX, our design for encrypted mempools, this represents a significant step towards solving the problem of MEV at the protocol level. In nearly every blockchain today, a single party ends up in control of each block: it decides which transactions get in, and can reorder them at will. MCP is the natural fix, but most recent designs pay for it with a separate aggregation phase, adding two extra communication rounds per block. Cadence makes the proposers part of consensus itself. Its fast path finalizes in an optimal three communication rounds, even when proposers are offline. Cadence also offers speculative finality, similar to MonadBFT, after just two rounds, revertible only if a proposer provably equivocated. In a simulation using estimated network delays between Monad mainnet's 200 globally distributed validators, finalization takes 219 ms on average, speculative finality 167 ms. Cadence pushes pipelining to the extreme: each block is proposed and finalized in its own independent consensus instance, without waiting on preceding blocks. The block interval then becomes a protocol parameter that can be arbitrarily small. At our initial target of 100 ms, a transaction waits on average just 50 ms to enter a proposal, and oracle prices, liquidations, and auctions can update every 100 ms. Cadence dynamically throttles the opening of new instances to bound the number of outstanding slots even during periods of network instability. When the network is healthy (under synchrony), a transaction included by an honest proposer can be neither dropped nor deferred (short-term censorship resistance), and no proposer can see the others' proposals in time to react (hiding). We prove both, together with safety and liveness under partial synchrony at the optimal 3f+1 fault bound. The Cadence protocol is modular: each module is simple on its own, and any of them can be swapped out without touching the rest. Cadence also builds on components already being deployed: proposals are disseminated as erasure-coded chunks over Deterministic RaptorCast, now rolling out on Monad, and validators vote on proposal digests, so voting does not wait for the full data to arrive. Start with the interactive tutorial: Full paper: Joint work by Kushal Babel, Fatima Elsheimy, Lioba Heimbach, Mohammad Mussadiq Jalalzai, Tobias Klenze, Jovan Komatovic, Jason Milionis, Mike Setrin, and Victor Shoup.

Category Labs

252,683 次观看 • 2 个月前

Transformer by hand ✍️ ~ 6 steps walkthrough below Open the hood of a transformer and the parts list is overwhelming: embeddings, positional encoding, attention weighting, self-attention, cross-attention, multi-head attention, layer norm, skip connections, softmax, linear, Nx, shifted right, query, key, value, masking. Which of those actually make the car run? Two of them. Attention weighting and the feed-forward network. Everything else is an enhancement to make it run faster and longer, which is how we got from a car to a truck, and to the word "large" in large language model. So I drew and calculated those two parts entirely by hand. Goal: push five features through one transformer block, filling in every cell yourself. 1. Given Five positions of input features, arriving from the previous block. 2. Attention matrix Let us feed all five features to a query-key module (QK) and read back an attention weight matrix, A. The details of that module are a post of their own. 3. Attention weighting We multiply the input features by A to get the attention weighted features, Z. Still five positions. The effect is to combine features *across positions*, horizontally: X1 becomes X1 + X2, X2 becomes X2 + X3, and so on. 4. First layer Let us feed all five weighted features into the first layer of the FFN. Multiply by the weights and biases. This time the combining happens *across feature dimensions*, vertically, and each feature grows from 3 numbers to 4. Note that every position goes through the same weight matrix. That is what "position-wise" means. 5. ReLU We cross out the negatives. They become zeros. 6. Second layer Let us bring it back down: 4 dimensions to 3. The output feeds the next block, which has a completely separate set of parameters, and the whole thing runs again. You have just calculated a transformer block by hand. ✍️ The takeaway: the two parts are doing two different jobs, and neither one alone is enough. Attention mixes *across positions*, so a feature can see its neighbours. The FFN mixes *across feature dimensions*, so each position can think about itself. Horizontal, then vertical. Then that pattern repeats N times, each block with its own separate set of weights. That is the Nx from the list up top, and that is what makes the transformer run. 💾 Save this post! #AIbyHand #Transformers #DeepLearning
0:12

Transformer by hand ✍️ ~ 6 steps walkthrough below Open the hood of a transformer and the parts list is overwhelming: embeddings, positional encoding, attention weighting, self-attention, cross-attention, multi-head attention, layer norm, skip connections, softmax, linear, Nx, shifted right, query, key, value, masking. Which of those actually make the car run? Two of them. Attention weighting and the feed-forward network. Everything else is an enhancement to make it run faster and longer, which is how we got from a car to a truck, and to the word "large" in large language model. So I drew and calculated those two parts entirely by hand. Goal: push five features through one transformer block, filling in every cell yourself. 1. Given Five positions of input features, arriving from the previous block. 2. Attention matrix Let us feed all five features to a query-key module (QK) and read back an attention weight matrix, A. The details of that module are a post of their own. 3. Attention weighting We multiply the input features by A to get the attention weighted features, Z. Still five positions. The effect is to combine features *across positions*, horizontally: X1 becomes X1 + X2, X2 becomes X2 + X3, and so on. 4. First layer Let us feed all five weighted features into the first layer of the FFN. Multiply by the weights and biases. This time the combining happens *across feature dimensions*, vertically, and each feature grows from 3 numbers to 4. Note that every position goes through the same weight matrix. That is what "position-wise" means. 5. ReLU We cross out the negatives. They become zeros. 6. Second layer Let us bring it back down: 4 dimensions to 3. The output feeds the next block, which has a completely separate set of parameters, and the whole thing runs again. You have just calculated a transformer block by hand. ✍️ The takeaway: the two parts are doing two different jobs, and neither one alone is enough. Attention mixes *across positions*, so a feature can see its neighbours. The FFN mixes *across feature dimensions*, so each position can think about itself. Horizontal, then vertical. Then that pattern repeats N times, each block with its own separate set of weights. That is the Nx from the list up top, and that is what makes the transformer run. 💾 Save this post! #AIbyHand #Transformers #DeepLearning

Tom Yeh

26,089 次观看 • 1 个月前

POLKADOT'S 2026 IS LOOKING GOOD Polkadot's $DOT has had a genuinely eventful start to 2026, and the catalysts are structural rather than speculative. Three developments stand out... (1) FIRST-EVER US SPOT ETF: 21Shares launched the first US spot Polkadot ETF on March 6, 2026, trading on Nasdaq under the ticker $TDOT, seeded with $11 million and carrying a 0.3% management fee. (2) HARD SUPPLY CAP: A tokenomics upgrade in March 2026 introduced a hard cap on DOT issuance, giving the token a max supply of 2.1 billion. (3) POLKADOT HUB IS LIVE: Polkadot Hub, the network's unified L1 application platform, is now live, simplifying dApp development by integrating core functionalities across system chains and supporting both EVM and the new RISC-V-based PVM. Despite a 90%+ price decline from its 2021 peak, Polkadot is still very much alive and kicking. The price hasn't caught up yet. The question is whether it will.
0:16

POLKADOT'S 2026 IS LOOKING GOOD Polkadot's $DOT has had a genuinely eventful start to 2026, and the catalysts are structural rather than speculative. Three developments stand out... (1) FIRST-EVER US SPOT ETF: 21Shares launched the first US spot Polkadot ETF on March 6, 2026, trading on Nasdaq under the ticker $TDOT, seeded with $11 million and carrying a 0.3% management fee. (2) HARD SUPPLY CAP: A tokenomics upgrade in March 2026 introduced a hard cap on DOT issuance, giving the token a max supply of 2.1 billion. (3) POLKADOT HUB IS LIVE: Polkadot Hub, the network's unified L1 application platform, is now live, simplifying dApp development by integrating core functionalities across system chains and supporting both EVM and the new RISC-V-based PVM. Despite a 90%+ price decline from its 2021 peak, Polkadot is still very much alive and kicking. The price hasn't caught up yet. The question is whether it will.

BSCN

43,483 次观看 • 4 个月前

🔥STRATEGY WILL BE THE WORLD'S MOST VALUABLE COMPANY🔥 Strategy bought OVER 56,000 Bitcoin in April. That number is so absurd people are psychologically incapable of processing it. Post-halving miners produce roughly 13,500 BTC per month. Strategy just bought about 4.1x an entire month of new miner supply in one month. Now run the simple monster math: Today: Strategy BTC stack: 818,334 BTC Bitcoin price: $76,196 Bitcoin NAV: $62.35B Assume Strategy keeps buying 56,000 BTC per month for 5 years. That is: ASSUMING STRC GROWTH TOTALLY STOPS (LOL) ~672,000 BTC per year ~3,360,000 BTC over 5 years Their stack goes from: 818,334 BTC to 4,178,334 BTC Now assume Bitcoin compounds at 25% CAGR. Bitcoin goes from: $76,196 to roughly: $232,532 So the Bitcoin NAV becomes: 4,178,334 BTC × $232,532 = roughly $971.5 BILLION Almost $1 TRILLION in Bitcoin NAV. And the funniest part? This model assumes no mNAV expansion. No premium insanity. No additional acceleration. No credit flywheel getting stronger. No market panic as everyone realizes Strategy is vacuuming Bitcoin off the planet like a publicly traded monetary black hole. Just: 56,000 BTC per month. 25% Bitcoin CAGR. 5 years. That’s it. Don't think they can accumulate that much Bitcoin at that low of a CAGR? Think the Bitcoin CAGR has to go higher? Cool. That only helps Strategy buy more Bitcoin. The bear case is basically: “Sure, they are absorbing multiples of new supply, building the largest corporate Bitcoin balance sheet in history, converting fiat capital markets into Bitcoin ownership, and compounding NAV at escape velocity, but have you considered that I am emotionally upset?” MSTR is becoming the most aggressive Bitcoin accumulation machine ever built. The fiat world is still modeling it like a tech stock with a weird treasury policy. GOOD LUCK.
0:14

🔥STRATEGY WILL BE THE WORLD'S MOST VALUABLE COMPANY🔥 Strategy bought OVER 56,000 Bitcoin in April. That number is so absurd people are psychologically incapable of processing it. Post-halving miners produce roughly 13,500 BTC per month. Strategy just bought about 4.1x an entire month of new miner supply in one month. Now run the simple monster math: Today: Strategy BTC stack: 818,334 BTC Bitcoin price: $76,196 Bitcoin NAV: $62.35B Assume Strategy keeps buying 56,000 BTC per month for 5 years. That is: ASSUMING STRC GROWTH TOTALLY STOPS (LOL) ~672,000 BTC per year ~3,360,000 BTC over 5 years Their stack goes from: 818,334 BTC to 4,178,334 BTC Now assume Bitcoin compounds at 25% CAGR. Bitcoin goes from: $76,196 to roughly: $232,532 So the Bitcoin NAV becomes: 4,178,334 BTC × $232,532 = roughly $971.5 BILLION Almost $1 TRILLION in Bitcoin NAV. And the funniest part? This model assumes no mNAV expansion. No premium insanity. No additional acceleration. No credit flywheel getting stronger. No market panic as everyone realizes Strategy is vacuuming Bitcoin off the planet like a publicly traded monetary black hole. Just: 56,000 BTC per month. 25% Bitcoin CAGR. 5 years. That’s it. Don't think they can accumulate that much Bitcoin at that low of a CAGR? Think the Bitcoin CAGR has to go higher? Cool. That only helps Strategy buy more Bitcoin. The bear case is basically: “Sure, they are absorbing multiples of new supply, building the largest corporate Bitcoin balance sheet in history, converting fiat capital markets into Bitcoin ownership, and compounding NAV at escape velocity, but have you considered that I am emotionally upset?” MSTR is becoming the most aggressive Bitcoin accumulation machine ever built. The fiat world is still modeling it like a tech stock with a weird treasury policy. GOOD LUCK.

Adam Livingston

61,582 次观看 • 4 个月前

Is Michael Saylor about to get a margin call? No. And the reason is more interesting than the rumor, because what he built instead may be harder to escape than one. A margin call needs a lender who can seize collateral when the price drops. Strategy has none. Its $6.7 billion in debt is convertible notes, the largest tranche due in 2029, with no loan-to-value trigger and no clause that lets anyone take a coin because Bitcoin fell. Saylor learned that in 2022, when he did have a collateralized loan and sweated a liquidation price, then rebuilt the structure so it could never happen again. On the literal question he is right, and the people calling for his liquidation this week do not understand what they see. But killing the fast death created a slow one almost nobody is pricing. To fund his buying, Saylor issued a mountain of perpetual preferred stock that pays a fixed dividend forever, near 11.5 percent, no matter where Bitcoin trades. That annual bill quadrupled from about $300 million in January to roughly $1.2 billion now, while the cash reserve that pays it fell 38 percent this year to near $1.4 billion, after the company spent $1.5 billion in May retiring debt. Put those two numbers together and you get the figure that actually matters, and it is not a Bitcoin price. It is a countdown. Dividend coverage, the time the cash can keep paying that bill, has collapsed from more than seven years in early 2026 to between ten and fourteen months, depending on whose math you use. Months, not years. The market is already pricing it, just not where the rumor is looking. That preferred stock is engineered to sit at $100. Last week it cracked to $82.50, a record 17.5 percent below par. That discount is investors quietly clocking the strain while the timeline screams about a margin call that cannot happen. There is a clean way out, and it is the one door the structure was built to keep shut. Restoring a safe two years of coverage takes about $2.8 billion, roughly double what Strategy holds, and the fastest path there is to sell Bitcoin. But selling crystallizes a $10.6 billion loss, breaks the never-sell promise that gives the stock its premium, and bleeds the very asset the machine exists to hoard. The exit and the wound are the same cut. He already brushed it, selling 32 coins on June 1 to cover a payment. Thirty-two against more than 847,000 is a rounding error in size and an earthquake in meaning, because the company that swore it would never sell, sold, to pay a dividend. And there is a second trigger almost no one has read, buried in the fine print. If Saylor ever simply skips a preferred payment to save cash, the missed amount compounds, the senior layer can ratchet its rate higher, a senior miss freezes payments to every junior layer beneath it, and after enough missed quarters those preferred holders can start taking board seats. No one seizes a coin. But control begins migrating to the people he owes. The clock does not just run down. It hands away the keys at the end. So the honest verdict is the one neither side is shouting. There is no margin call and no imminent bankruptcy. The structure protects him exactly as designed. What it cannot protect him from is a fixed bill that grows while the cash shrinks, where every exit deepens the hole. Sell Bitcoin and break the story. Issue stock into a price near its lowest since 2024 and punish your holders. Skip the dividend and start losing the company by the boardroom. Saylor did not escape the margin call. He traded a cliff for a clock. A cliff takes you in an afternoon and a stranger pulls the trigger. This clock takes months, and at the end the trigger is pulled by the only two forces he swore would never touch it, his own hand, or the people he owes. The rumor asks whether someone is about to call his loan. The real question is how many months he can keep paying before he has to sell the dream, dilute the believers, or hand over the board to keep the lights on.
0:10

Is Michael Saylor about to get a margin call? No. And the reason is more interesting than the rumor, because what he built instead may be harder to escape than one. A margin call needs a lender who can seize collateral when the price drops. Strategy has none. Its $6.7 billion in debt is convertible notes, the largest tranche due in 2029, with no loan-to-value trigger and no clause that lets anyone take a coin because Bitcoin fell. Saylor learned that in 2022, when he did have a collateralized loan and sweated a liquidation price, then rebuilt the structure so it could never happen again. On the literal question he is right, and the people calling for his liquidation this week do not understand what they see. But killing the fast death created a slow one almost nobody is pricing. To fund his buying, Saylor issued a mountain of perpetual preferred stock that pays a fixed dividend forever, near 11.5 percent, no matter where Bitcoin trades. That annual bill quadrupled from about $300 million in January to roughly $1.2 billion now, while the cash reserve that pays it fell 38 percent this year to near $1.4 billion, after the company spent $1.5 billion in May retiring debt. Put those two numbers together and you get the figure that actually matters, and it is not a Bitcoin price. It is a countdown. Dividend coverage, the time the cash can keep paying that bill, has collapsed from more than seven years in early 2026 to between ten and fourteen months, depending on whose math you use. Months, not years. The market is already pricing it, just not where the rumor is looking. That preferred stock is engineered to sit at $100. Last week it cracked to $82.50, a record 17.5 percent below par. That discount is investors quietly clocking the strain while the timeline screams about a margin call that cannot happen. There is a clean way out, and it is the one door the structure was built to keep shut. Restoring a safe two years of coverage takes about $2.8 billion, roughly double what Strategy holds, and the fastest path there is to sell Bitcoin. But selling crystallizes a $10.6 billion loss, breaks the never-sell promise that gives the stock its premium, and bleeds the very asset the machine exists to hoard. The exit and the wound are the same cut. He already brushed it, selling 32 coins on June 1 to cover a payment. Thirty-two against more than 847,000 is a rounding error in size and an earthquake in meaning, because the company that swore it would never sell, sold, to pay a dividend. And there is a second trigger almost no one has read, buried in the fine print. If Saylor ever simply skips a preferred payment to save cash, the missed amount compounds, the senior layer can ratchet its rate higher, a senior miss freezes payments to every junior layer beneath it, and after enough missed quarters those preferred holders can start taking board seats. No one seizes a coin. But control begins migrating to the people he owes. The clock does not just run down. It hands away the keys at the end. So the honest verdict is the one neither side is shouting. There is no margin call and no imminent bankruptcy. The structure protects him exactly as designed. What it cannot protect him from is a fixed bill that grows while the cash shrinks, where every exit deepens the hole. Sell Bitcoin and break the story. Issue stock into a price near its lowest since 2024 and punish your holders. Skip the dividend and start losing the company by the boardroom. Saylor did not escape the margin call. He traded a cliff for a clock. A cliff takes you in an afternoon and a stranger pulls the trigger. This clock takes months, and at the end the trigger is pulled by the only two forces he swore would never touch it, his own hand, or the people he owes. The rumor asks whether someone is about to call his loan. The real question is how many months he can keep paying before he has to sell the dream, dilute the believers, or hand over the board to keep the lights on.

Shanaka Anslem Perera ⚡

58,558 次观看 • 2 个月前

Meanwhile, Putin arrived at an aircraft plant in Irkutsk and is delighted with the long-suffering "Russian-made" MS-21 airliner. They showed him the supposed production of the first 18 aircraft, but it is just dust in the eyes of the dictator. The MS-21 is more than 10 years behind schedule. Development costs have ballooned from 150 billion to over 1 trillion rubles, and the unit cost has roughly tripled. Due to sanctions, the plane gained 5–6 tons in weight, lost half its range, was shortened, and now carries almost 30% fewer passengers. By mid-2026, only a third of the required test flights with Russian components had been completed.
0:17

Meanwhile, Putin arrived at an aircraft plant in Irkutsk and is delighted with the long-suffering "Russian-made" MS-21 airliner. They showed him the supposed production of the first 18 aircraft, but it is just dust in the eyes of the dictator. The MS-21 is more than 10 years behind schedule. Development costs have ballooned from 150 billion to over 1 trillion rubles, and the unit cost has roughly tripled. Due to sanctions, the plane gained 5–6 tons in weight, lost half its range, was shortened, and now carries almost 30% fewer passengers. By mid-2026, only a third of the required test flights with Russian components had been completed.

Visegrád 24

240,792 次观看 • 1 个月前

NEWS: Tesla's fleet has now driven 12 billion miles on Full Self-Driving The count passed 12 billion this week, just a month after hitting 11 billion on June 9. About 4.5 billion of those miles were on city streets. The pace is speeding up fast. The fleet now adds around 30 million miles a day, up from 14 million a day at the start of the year. Each new billion is arriving quicker than the last. Why this number matters. Elon Musk has said Tesla needs roughly 10 billion miles of real driving data to make self-driving safe enough to run unsupervised. The fleet is now well past that and the data trains the system to handle rare, tricky situations. Tesla's own safety report says cars on FSD have one major crash every 5.3 million miles, versus one every 660,000 for the average US driver. FSD is still a supervised system. The driver has to stay attentive and ready to take over. (Source: Tesla FSD Safety Report, July 17)
0:17

NEWS: Tesla's fleet has now driven 12 billion miles on Full Self-Driving The count passed 12 billion this week, just a month after hitting 11 billion on June 9. About 4.5 billion of those miles were on city streets. The pace is speeding up fast. The fleet now adds around 30 million miles a day, up from 14 million a day at the start of the year. Each new billion is arriving quicker than the last. Why this number matters. Elon Musk has said Tesla needs roughly 10 billion miles of real driving data to make self-driving safe enough to run unsupervised. The fleet is now well past that and the data trains the system to handle rare, tricky situations. Tesla's own safety report says cars on FSD have one major crash every 5.3 million miles, versus one every 660,000 for the average US driver. FSD is still a supervised system. The driver has to stay attentive and ready to take over. (Source: Tesla FSD Safety Report, July 17)

Muskonomy

34,557 次观看 • 1 个月前

THE CLOUD BILL WAS $14,000 A MONTH. EVERY MONTH. JUST FOR GPU ACCESS. No ownership. No hardware. Just renting someone else's chips and watching the invoice grow. So this startup did the math and bought 1,000 Mac Mini M4s instead. $599 each. One-time. Total: ~$599,000 upfront. Sounds insane. Until you do the math the other way. $14K/mo in cloud GPUs is $168K a year. In 3.5 years you've burned through $599K and own nothing. The meter just keeps running. These guys own the hardware. Same output. Less power draw. And after year 4, every month of compute is basically free. The M4 was never built for data centers. But its performance per watt is so good that someone looked at a rack of a thousand of them and thought "why not." That's when it clicks. The AI infrastructure race isn't about who has the biggest GPU anymore. It's about who figured out they were overpaying for one. Would you make this switch?
0:13

THE CLOUD BILL WAS $14,000 A MONTH. EVERY MONTH. JUST FOR GPU ACCESS. No ownership. No hardware. Just renting someone else's chips and watching the invoice grow. So this startup did the math and bought 1,000 Mac Mini M4s instead. $599 each. One-time. Total: ~$599,000 upfront. Sounds insane. Until you do the math the other way. $14K/mo in cloud GPUs is $168K a year. In 3.5 years you've burned through $599K and own nothing. The meter just keeps running. These guys own the hardware. Same output. Less power draw. And after year 4, every month of compute is basically free. The M4 was never built for data centers. But its performance per watt is so good that someone looked at a rack of a thousand of them and thought "why not." That's when it clicks. The AI infrastructure race isn't about who has the biggest GPU anymore. It's about who figured out they were overpaying for one. Would you make this switch?

Framez

18,004 次观看 • 1 个月前

🚨 THIS IS HOW THE NEXT GLOBAL CRISIS BEGINS, AND IT IS ALREADY UNDERWAY IN THE ONE MARKET NOBODY WATCHES Three weeks ago I warned about a debt crisis building. Since then every number moved the wrong way. Here is the truth almost nobody wants to hear. The S&P 500, Bitcoin, all of crypto, they are dust compared to the bond market. The global bond market sets the price of money itself. Every stock, every coin, every mortgage, every government is built on top of it. When it moves, everything moves. And right now, it is cracking on three fronts at once. THE UNITED STATES → 30Y yield: 5.31%, a 19-YEAR high → 10Y: 4.73%, closing in on 5% → Levels last seen right before 2008 → Foreign buyers pulled $72B out in a single month America's biggest lenders are walking away. CHINA → Cut its US Treasury holdings to $633B, an 18-YEAR low → The lowest since September 2008 For years China was one of the biggest buyers of US debt. Now it is doing the opposite, steadily backing away from funding America and the rest of the world. The buyer that helped keep borrowing cheap is gone. JAPAN → Near-zero rates for THREE DECADES, now ending → 10Y yield hit 3%, highest since 1996 → 30Y broke above 4% When Japanese bonds pay again, Japanese money stops funding the world and stays home. Another giant buyer disappears. EUROPE → France's 10Y at 4.10%, highest since 2009 → Germany's 30Y at 3.73%, highest since 2011 → UK's 30Y gilt at 5.85% France is the weak link. Italy must refinance debt worth 17% of its GDP this year. Every major government on earth is paying more to borrow, at the same time, for the same reasons. The trigger could be anything. But the weakest point in the whole system is Japan. Three decades of cheap money unwinding at once is the crack that could split the entire foundation. Watch the bond market first. Everything else is a sideshow. 12 years in these markets. This is what I do. Follow me and turn notifications on.
0:10

🚨 THIS IS HOW THE NEXT GLOBAL CRISIS BEGINS, AND IT IS ALREADY UNDERWAY IN THE ONE MARKET NOBODY WATCHES Three weeks ago I warned about a debt crisis building. Since then every number moved the wrong way. Here is the truth almost nobody wants to hear. The S&P 500, Bitcoin, all of crypto, they are dust compared to the bond market. The global bond market sets the price of money itself. Every stock, every coin, every mortgage, every government is built on top of it. When it moves, everything moves. And right now, it is cracking on three fronts at once. THE UNITED STATES → 30Y yield: 5.31%, a 19-YEAR high → 10Y: 4.73%, closing in on 5% → Levels last seen right before 2008 → Foreign buyers pulled $72B out in a single month America's biggest lenders are walking away. CHINA → Cut its US Treasury holdings to $633B, an 18-YEAR low → The lowest since September 2008 For years China was one of the biggest buyers of US debt. Now it is doing the opposite, steadily backing away from funding America and the rest of the world. The buyer that helped keep borrowing cheap is gone. JAPAN → Near-zero rates for THREE DECADES, now ending → 10Y yield hit 3%, highest since 1996 → 30Y broke above 4% When Japanese bonds pay again, Japanese money stops funding the world and stays home. Another giant buyer disappears. EUROPE → France's 10Y at 4.10%, highest since 2009 → Germany's 30Y at 3.73%, highest since 2011 → UK's 30Y gilt at 5.85% France is the weak link. Italy must refinance debt worth 17% of its GDP this year. Every major government on earth is paying more to borrow, at the same time, for the same reasons. The trigger could be anything. But the weakest point in the whole system is Japan. Three decades of cheap money unwinding at once is the crack that could split the entire foundation. Watch the bond market first. Everything else is a sideshow. 12 years in these markets. This is what I do. Follow me and turn notifications on.

DeFi_Machine

322,129 次观看 • 20 天前