Crypto World
What is basis trading? Cash-and-carry arbitrage
Basis trading is a market-neutral strategy that profits from the price gap between spot Bitcoin and its futures contracts. It is the reason hedge funds hold billions in Bitcoin ETFs without betting on the price going up.
Summary
- Basis trading, also called cash-and-carry arbitrage, involves buying an asset in the spot market and simultaneously selling a futures contract on the same asset, locking in the price difference as profit regardless of which direction the market moves.
- The strategy became the dominant institutional play in crypto after spot Bitcoin ETFs launched in January 2024, with hedge funds using ETF shares as the spot leg and CME futures as the short leg to capture annualized yields that have ranged from 5% to more than 20%.
- The “basis” is the difference between the futures price and the spot price. In crypto markets, futures almost always trade at a premium to spot because leveraged traders are willing to pay more for exposure without holding the underlying asset. That premium is what basis traders harvest.
- Basis trading is not directional. The trader does not profit from Bitcoin going up or down. The profit comes exclusively from the convergence of the futures price and the spot price as the contract approaches expiration, a mathematical certainty barring exchange default.
- The strategy carries risks including margin calls on the short futures leg during sharp rallies, counterparty risk on the futures exchange, liquidity risk if the ETF shares cannot be sold quickly, and opportunity cost if Bitcoin rallies significantly while the position is locked.
The most widely repeated misunderstanding about Bitcoin ETF inflows is that they represent bullish bets on the price. Many of them do. But a significant share of the billions flowing into spot Bitcoin ETFs comes from hedge funds and trading firms that are completely indifferent to whether Bitcoin goes up or down. They are running basis trades, and the only number they care about is the spread between spot and futures.
This guide explains how the trade works mechanically, why crypto markets offer higher basis yields than traditional commodities, what risks the strategy carries, and how to evaluate whether the current basis is worth capturing. Understanding the basis trade is also essential for interpreting ETF flow data, futures open interest, and funding rate charts, because each of these metrics is heavily influenced by basis trading activity that is often misread as directional conviction.
How the basis trade works step by step
The mechanics are straightforward once the terminology is clear. A basis trade requires two simultaneous positions: a long position in the spot market and a short position in the futures market for the same asset and the same notional amount.
Step one: the trader buys $1 million worth of Bitcoin at the current spot price. In the ETF era, this typically means purchasing shares of a spot Bitcoin ETF such as BlackRock IBIT or Fidelity FBTC, which track Bitcoin’s price through direct holdings of the asset. The spot ETF creation and redemption mechanism ensures that ETF shares trade close to the net asset value of the underlying Bitcoin.
Step two: the trader simultaneously sells $1 million worth of Bitcoin futures on a regulated exchange, most commonly the CME. The futures contract will expire on a set date, typically monthly or quarterly.
Step three: the trader holds both positions until the futures contract expires. At expiration, the futures price converges with the spot price by definition, because the contract settles against the actual spot price. The difference between the price at which the futures were sold and the price at which they converge is the trader’s profit.
If Bitcoin was trading at $100,000 spot and the one-month futures contract was trading at $101,500, the basis is $1,500 or 1.5% for one month. Annualized, that is approximately 18%. The trader collects that 1.5% regardless of whether Bitcoin finishes the month at $80,000 or $120,000, because the gains on one leg offset the losses on the other.
Why crypto basis is higher than traditional markets
In traditional commodity markets, the basis on oil, gold, or agricultural futures typically runs between 1% and 5% annualized. In crypto markets, the annualized basis has historically ranged from 5% to more than 25%, with spikes above 40% during periods of extreme bullish sentiment. During the bull run of late 2024 and early 2025, the CME Bitcoin front-month basis routinely exceeded 15% annualized, a yield that no comparable fixed-income instrument could match at the time.
The reason is structural. Crypto futures markets are dominated by leveraged speculators who want long exposure without holding the underlying asset. This persistent demand for long futures pushes the futures price above the spot price, creating what traders call contango. The steeper the contango, the wider the basis, and the more profitable the cash-and-carry trade becomes.
Three factors keep crypto basis elevated compared to traditional markets. First, crypto markets trade around the clock every day of the year, which means funding costs and leverage demand never pause. The New York Mercantile Exchange closes on weekends. Binance and Bybit do not. Continuous trading means continuous demand for leverage, which translates to a persistently elevated premium on futures.
Second, the margining requirements on crypto futures are higher than on traditional commodity futures, which means the cost of maintaining leveraged positions is higher, and that cost gets priced into the futures premium. CME Bitcoin futures require initial margin around 40%, compared to roughly 5% to 10% for crude oil or gold. The higher the margin requirement, the more capital a leveraged long must deploy, and the more premium they are willing to accept.
Third, retail participation in crypto futures is proportionally larger than in traditional markets, and retail traders tend to be net long and willing to pay higher premiums for leveraged upside. On offshore exchanges, it is common to see 50x or 100x leverage on Bitcoin perpetual contracts. These highly leveraged longs create enormous demand for the other side of the trade, and the basis is the price the market pays to satisfy that demand.
The perpetual futures funding rate is a related concept. Perpetual contracts do not expire, so there is no natural convergence date. Instead, exchanges use a funding rate mechanism where longs pay shorts (or vice versa) every eight hours to keep the perp price anchored to spot. When funding rates are positive and elevated, it signals the same demand imbalance that drives the basis on dated futures. During sustained bull markets, cumulative funding payments can exceed 30% annualized, making the perp funding trade even more lucrative than the dated futures version.
The ETF basis trade: how institutions do it
Before spot Bitcoin ETFs launched in January 2024, running a basis trade required holding actual Bitcoin on an exchange or with a custodian. This introduced counterparty risk, custody complexity, and regulatory ambiguity that kept most institutional capital away.
The ETF changed the calculation entirely. A hedge fund can now buy IBIT shares through a prime broker, short CME Bitcoin futures through the same prime broker, and report both positions on a single balance sheet with no direct crypto custody. The trade settles in dollars, clears through regulated infrastructure, and fits within existing risk frameworks.
SEC 13F filings have revealed the scale of this activity. Millennium Management, Citadel, Point72, and dozens of other multi-strategy hedge funds disclosed large IBIT positions alongside corresponding CME futures shorts. These are not Bitcoin bulls. They are arbitrageurs harvesting the basis, and their ETF flow activity creates the paradox of billions in ETF inflows that carry zero directional conviction.
The institutional version of the trade typically targets annualized returns of 8% to 15% with minimal drawdown risk. For a fund that can borrow at 5%, a 12% annualized basis produces 7% of alpha on what is effectively a market-neutral position. At institutional scale, that is an attractive risk-adjusted return.
The scale of institutional basis trading explains a pattern that confuses many retail observers. ETF inflows can surge on a day when Bitcoin’s price barely moves, and they can remain strong during periods of sideways trading. This happens because basis traders are responding to futures premium levels, not to price direction. A widening basis attracts more capital into the trade regardless of whether Bitcoin is trending up, down, or sideways. Conversely, when the basis compresses below the cost of capital, institutional ETF flows can dry up even during a rally, because the arbitrage no longer pays.
The perpetual funding rate trade
The dated futures basis trade has a cousin: the perpetual funding rate trade. Instead of buying spot and shorting a dated future, the trader buys spot and shorts a perpetual contract on a crypto exchange such as Binance, Bybit, or Hyperliquid.
The profit mechanism is different. There is no expiration date and no convergence event. Instead, the trader collects funding payments every eight hours when the funding rate is positive. Positive funding means longs are paying shorts, which means the trader holding the short perp leg receives payments continuously.
The advantage of the funding rate trade is flexibility. The trader can enter and exit at any time without waiting for contract expiration. The disadvantage is unpredictability. Funding rates can turn negative during bearish periods, at which point the short leg starts costing money instead of earning it. The trader must monitor rates actively and be prepared to unwind when the trade stops paying.
The funding rate version also carries higher counterparty risk because it typically involves unregulated offshore exchanges. The CME basis trade, by contrast, clears through a regulated clearinghouse, which is why institutional capital overwhelmingly prefers the dated futures version.
A hybrid approach exists for traders who want the flexibility of perpetuals with reduced counterparty risk. Some traders hold their spot leg in a self-custodied wallet or on a regulated exchange and run the short perp leg on a decentralized perpetual exchange such as Hyperliquid or dYdX. The smart contract handles margin and settlement without an intermediary, which removes the centralized exchange failure risk. The tradeoff is that decentralized perp venues sometimes have lower liquidity and wider spreads than their centralized counterparts, which increases execution costs.
The arithmetic: when the trade pays and when it does not
The profitability of a basis trade depends on four numbers: the current basis spread, the cost of capital, the margin requirements, and the holding period.
Consider a concrete example. Bitcoin spot is at $100,000. The three-month CME futures contract trades at $104,000. The annualized basis is approximately 16%. The trader buys $10 million in IBIT shares and shorts $10 million in CME futures.
If the trader’s cost of capital is 5% (prime broker financing), the net yield is 11% annualized. Over three months, that produces approximately $275,000 in profit on $10 million of notional, with near-zero directional risk.
But the arithmetic changes if the basis compresses. If Bitcoin enters a bearish period and futures flip to backwardation (futures below spot), there is no basis to capture and the trade produces a loss. Historically, crypto futures have been in contango approximately 85% of the time, which is why the trade has been consistently profitable over multi-year periods.
The arithmetic also changes with margin. CME Bitcoin futures require initial margin of roughly 40% of notional. If Bitcoin rallies sharply, the short futures leg generates unrealized losses that require additional margin. A 20% rally on a $10 million short futures position creates $2 million in margin calls. The trader must have sufficient liquidity to meet those calls without unwinding the position, because unwinding the short leg while keeping the long leg converts a market-neutral trade into a directional long that may then reverse.
This margin dynamic is the single most common cause of basis trade failure. During the rally from $60,000 to $73,000 in March 2024, several smaller funds were forced to close their short futures legs because they could not meet margin calls. Their IBIT positions, no longer hedged, became naked longs at exactly the moment the rally paused and reversed. The trade that was designed to be market neutral became a directional loss because the fund did not hold enough reserve capital to survive the short-term drawdown on the short leg.
Roll cost is another factor that reduces realized returns. When a dated futures contract approaches expiration, the trader must close the expiring short and open a new short in the next contract month. This roll carries transaction costs, including commissions, the bid-ask spread on both the closing and opening legs, and potential slippage if the roll happens during a volatile session. For quarterly rolls on CME Bitcoin futures, these costs typically consume 0.1% to 0.3% of notional per roll, which can reduce the annualized yield by one to two percentage points.
What this does not cover
This guide does not cover crypto arbitrage strategies beyond the cash-and-carry trade, such as triangular arbitrage, cross-exchange arbitrage, or statistical arbitrage. It does not cover options-based strategies that use the basis as an input, such as calendar spreads or volatility arbitrage. It does not cover the tax treatment of basis trades, which varies significantly by jurisdiction and depends on whether the spot leg is held as a security (ETF shares) or as property (direct cryptocurrency). It does not explain how to execute the trade on specific platforms, because execution details vary by exchange and broker and change frequently.
Practical checks before entering a basis trade
Check the current annualized basis. Platforms such as Coinglass, Laevitas, and The Block publish real-time annualized basis for CME and major exchange futures. If the annualized basis is below your cost of capital, the trade does not pay.
Check open interest on the contract you plan to short. Low open interest means the contract is illiquid, which widens the bid-ask spread and increases the cost of entry and exit. CME Bitcoin front-month contracts typically have sufficient liquidity for institutional-sized trades. Back-month contracts may not.
Check your margin buffer. Calculate the maximum drawdown your short leg can sustain before triggering a margin call. A common rule of thumb is to hold enough reserve capital to absorb a 30% to 40% rally without needing to unwind. If you cannot meet margin calls in a rally, the trade can turn from market-neutral to forced liquidation.
Check the funding rate if using perpetual contracts. Look at the 30-day average funding rate, not the current snapshot. A single elevated snapshot can be an anomaly. The 30-day average tells you whether the trade is structurally paying.
Check counterparty risk. On CME, your counterparty risk is the clearinghouse. On an offshore exchange, your counterparty risk is the exchange itself. If the exchange goes down, your short leg disappears and you are left with a naked long position in a potentially falling market.
Is basis trading risk free?
No. Basis trading is often described as low risk, not zero risk. The primary risks are margin calls on the short leg during sharp rallies, counterparty default on the futures exchange, liquidity risk if positions cannot be unwound at expected prices, and the possibility that the basis turns negative during bearish periods. The “risk free” label comes from the mathematical certainty that futures converge to spot at expiration, but the path between entry and expiration can involve significant mark-to-market losses on one leg that must be financed.
How much capital do I need to start a basis trade?
The minimum depends on the venue. CME Bitcoin futures have a contract size of five Bitcoin (approximately $500,000 at $100,000 per coin), which makes the standard contract unsuitable for retail traders. CME Micro Bitcoin futures (one-tenth of one Bitcoin) have lower notional requirements. On crypto-native exchanges, perpetual contracts can be opened with as little as a few hundred dollars, though the counterparty risk is correspondingly higher.
Why do hedge funds buy Bitcoin ETFs if they are not bullish?
Because the ETF is the cheapest and most operationally simple way to hold the spot leg of a basis trade. The hedge fund profits from the spread between the ETF price and the futures price, not from Bitcoin appreciation. The ETF position is fully hedged by the short futures position.
What happens to the basis trade when Bitcoin crashes?
The spot leg loses value, but the short futures leg gains an approximately equal amount. The net profit or loss is determined by the basis, not by the direction of Bitcoin. However, if the crash is severe enough to push futures into backwardation, the basis turns negative and the trade loses money until contango resumes.
Can I run a basis trade with Ethereum or other cryptocurrencies?
Yes. Basis trades can be executed on any asset with liquid spot and futures markets. Ethereum has an active basis on CME futures, and the launch of spot Ethereum ETFs created the same institutional playbook that IBIT enabled for Bitcoin. Solana, XRP, and other major cryptocurrencies have basis on offshore exchanges, though liquidity is lower and counterparty risk is higher. The general rule is that the more liquid the spot and futures markets, the tighter the execution costs and the more reliable the basis capture.
What is the difference between basis trading and funding rate farming?
Basis trading uses dated futures that expire on a set date, and the profit comes from the convergence of futures to spot at expiration. Funding rate farming uses perpetual contracts that never expire, and the profit comes from collecting funding payments every eight hours. The economic logic is similar, but the risk profiles differ because perpetual funding rates can fluctuate rapidly.
How do I calculate the annualized basis?
Take the futures premium as a percentage of the spot price, then multiply by (365 divided by the number of days until expiration). If spot is $100,000, futures are $102,000, and the contract expires in 60 days, the premium is 2% and the annualized basis is 2% multiplied by (365/60), which equals approximately 12.2%.
Does basis trading affect Bitcoin’s price?
Not directly, because basis trades are market neutral. The spot buying and futures selling roughly offset each other in terms of price impact. However, large-scale basis trading can increase liquidity in both spot and futures markets, which can reduce volatility. The ETF inflows driven by basis traders also increase the total assets under management of Bitcoin ETFs, which some analysts interpret as a demand signal even though the underlying motivation is arbitrage. The unwinding of basis trades can have a more noticeable effect. If basis traders close their positions in bulk during a period of low liquidity, the simultaneous selling of ETF shares and buying back of futures can create short-term price dislocations.
Disclaimer
This article is for informational purposes only and does not constitute financial, investment, or trading advice. Basis trading involves risks including margin calls, counterparty default, and potential loss of capital. Past performance of basis spreads does not guarantee future results. Always conduct your own research and consult a qualified financial advisor before making investment decisions. Information accurate as of August 6, 2026.
Morgan Stanley Investment Management launched the Morgan Stanley Ethereum Trust and Morgan Stanley Solana Trust on NYSE Arca, spot exchange-traded products that will stake portions of their holdings, the firm said in a press release published Tuesday. The launch deepens the first crypto ETP… Read the full story at The Defiant
Only 150 unique venture capital firms have participated in crypto funding rounds in July, the lowest monthly count since November 2020, according to CryptoRank data through July 28. The figure captures how far the investor base has compressed since the last bull market: at the May 2022 peak, 1,177… Read the full story at The Defiant
Bitcoin miner Marathon Digital Holdings, commonly known as MARA, reported a sharp swing from profit to a large net loss in the second quarter of 2026, even as it achieved its highest quarterly Bitcoin production in more than a year. The company’s results underscore how tightly miners remain tied to Bitcoin’s market price—especially when the accounting reflects changes in the fair value of Bitcoin held on balance sheets.
In its Q2 2026 SEC Form 10-Q, MARA said it recorded a net loss of $611.3 million, or $1.60 per diluted share, compared with net income of $808.2 million, or $1.84 per diluted share, in the second quarter of 2025. According to the filing, MARA mined 2,422 Bitcoin in the quarter, up 3% year over year, but that increase was more than offset by a 28% decline in the average Bitcoin price.
Key takeaways
- MARA posted a $611.3 million Q2 net loss, with the decline largely attributed to changes in the value of its Bitcoin holdings.
- Bitcoin production rose to 2,422 BTC in Q2 2026, but the revenue impact was overwhelmed by a 28% drop in the average Bitcoin price.
- As of June 30, MARA held 35,577 Bitcoin valued at $2.1 billion, placing it fourth among public Bitcoin holders cited in the company’s reporting context.
- The company used the quarter to restructure its power portfolio and capital position, while pushing further into AI and high-performance computing (HPC) infrastructure.
- MARA is targeting at least two AI/HPC data center lease signings by year-end, alongside additional Texas and Ohio expansion plans.
Profit-to-loss driven by Bitcoin’s price and holding valuation
The headline shift in MARA’s second-quarter performance is stark: profit in Q2 2025 gave way to a net loss in Q2 2026. In its 10-Q, MARA attributes the swing primarily to the impact of Bitcoin price movement on the accounting value of Bitcoin it holds, rather than to a deterioration in mining output.
That distinction matters for how investors interpret miner fundamentals. Production volumes increased, but the company’s overall earnings were pressured by weaker realized economics tied to Bitcoin’s price environment. Put simply, even better operational throughput did not translate into higher net earnings when the fair-value effects and average pricing moved against the company.
During an earnings call on Thursday, MARA CFO Salman Khan said, according to the company’s remarks, that “Two things defined Q2 for MARA. Bitcoin prices created a challenging revenue environment [and] we used the quarter to fundamentally transform our power portfolio and capital structure.”
Higher output, weaker average price
MARA’s mining performance in Q2 2026 was comparatively strong on the operational side. The company mined 2,422 Bitcoin, about 3% more than the year-ago quarter. However, the average Bitcoin price fell 28% over the same comparison period, which directly undermined revenue tied to the mined BTC and other Bitcoin-linked line items.
This is a recurring tension in the miner model: when BTC prices move lower, output growth can be muted by pricing and valuation effects. MARA’s quarter illustrates that point—production strength alone was not enough to counteract the market-driven decline in average pricing.
Beyond the mining figures, the company’s Bitcoin balance sheet also remained significant. As of June 30, MARA reported total holdings of 35,577 Bitcoin with a total fair value of $2.1 billion, reflecting both continued treasury accumulation and the sensitivity of the financial statements to BTC valuation changes. In that snapshot, the company was described as the fourth-largest public Bitcoin holder after Strategy, Twenty One Capital and Metaplanet.
MARA presses ahead with AI and HPC infrastructure deals
While mining remains central to MARA’s business, the company continues to frame its longer-term growth around expanding computational infrastructure for AI and high-performance workloads. Earlier in 2026, it acquired a majority stake in Exaion SaS, which operates high-performance computing data centers and secure cloud and AI infrastructure.
MARA also moved to accelerate data center development through partnerships. In February, it announced a strategic partnership with Starwood Capital Group and Starwood Digital Ventures aimed at enabling conversion of select MARA sites to meet demand from “enterprise, hyperscale and AI customers.”
On Thursday, MARA reiterated that it is pursuing near-term commercial milestones tied to those plans. The company said it is targeting at least two AI/HPC lease signings by year-end. CEO Fred Thiel said, according to the earnings call, that lease discussions are progressing across multiple sites and that MARA remains confident it can sign at least two leases before year-end.
Texas land plans and Ohio energy acquisition expand the runway
MARA’s infrastructure buildout includes both new land and additional power resources. In July, the company agreed to acquire a 1,200-acre powered site in Matagorda County, Texas, with expected access to up to 2 gigawatts of grid capacity by April 2028. MARA said the site is intended for AI and HPC workloads as well as Bitcoin mining.
The company also continues to pursue power capacity through a pending acquisition of Long Ridge Energy & Power in Ohio. MARA described the $1.5 billion deal as a potential source of up to 600 megawatts of AI and critical-IT load over time, indicating that it views energy access as a key enabler for both traditional mining operations and new revenue streams linked to enterprise computing.
Mining still the core—AI described as a complement, not a replacement
In a shareholder letter released alongside its quarterly results, Thiel said that Bitcoin mining remains the foundation of MARA’s business and that the cash flow generated by mining will continue to support other investments.
He also pushed back on the notion that MARA is shifting away from mining. “Ultimately, we do not view Bitcoin mining and AI infrastructure as competing businesses,” Thiel said, according to the letter. He further emphasized a capital allocation principle focused on deploying each megawatt into what he described as its highest-value application—sometimes mining in certain markets, and in others AI infrastructure, sovereign cloud, or enterprise computing.
For readers tracking MARA, the key question is how quickly these AI/HPC efforts can contribute stable cash flows that are less dependent on Bitcoin’s spot price. Near-term, the company’s targets—like at least two AI/HPC lease signings by year-end—will offer a clearer datapoint on whether the operational transformation hinted at in Q2 can translate into measurable commercial traction. Investors will also want to watch how future quarters reflect both mining output and the impact of Bitcoin price moves on the valuation of holdings, since that remains the dominant factor in the recent earnings swing.
Bitcoin climbed above $64,400 after the Federal Reserve held its benchmark rate at 3.50%-3.75% on Wednesday, then gave back the move when Chair Kevin Warsh opened his press conference by saying, "There is no soft inflation target." BTC was trading just under $64,000 at press time, up about 1% over… Read the full story at The Defiant
Senate Majority Leader John Thune reportedly confirmed that the chamber was “punting” the vote until September.
Lightning Labs, the company behind Lightning Network developer tools, announced a dedicated website for L402 on Wednesday, positioning the protocol as a rail for AI agents to pay in bitcoin. "With L402, agents can pay with bitcoin and natively authenticate. No accounts, no intermediaries, no humans… Read the full story at The Defiant
Uber Technologies (UBER) stock slid Wednesday after the tech giant offered mixed second-quarter results. The ride-hailing and food-delivery company’s bookings and adjusted profit forecast missed estimates. The San Francisco-based Uber earned an adjusted 81 cents per share for the June-ended quarter, up 35% from a year earlier. That beat the 80 cents per share that analysts polled by FactSet were forecasting.…
Copyright ©2026 Investor’s Business Daily, LLC. All rights reserved. 87990cbe856818d5eddac44c7b1cdeb8
Real-world assets are moving from the “issuance” phase into onchain utility, with activity on decentralized finance platforms increasingly tied to tokenized yield and collateral rather than general DeFi momentum. According to a joint report from CoinShares and Token Terminal released Thursday, RWA deposits across DeFi more than tripled year over year to $7.4 billion in Q2 2026, even as total DeFi deposits declined by about 15%.
CoinShares CEO Jean-Marie Mognetti said the split between rising RWA balances and weakening broader DeFi inflows points to demand rooted in functional use cases. In his view, when a category expands while its host ecosystem softens, it is typically because users want financial utility—not because market conditions are broadly improving.
Key takeaways
- RWA deposits in DeFi surged to $7.4 billion in Q2 2026, while overall DeFi deposits fell about 15% year over year.
- Yield-bearing stablecoins and tokenized Treasury products are currently the largest RWA categories used onchain, including Sky Protocol’s sUSDS and BlackRock’s BUIDL.
- RWA spot trading on decentralized exchanges rose about 220% year over year, even though overall DEX volumes dropped roughly 70%.
- Gold-backed tokens and yield-oriented dollar products are central to RWA activity, with traders responding to price moves and yield strategies.
- RWA exposure is extending into derivatives: RWA perpetual futures volumes have continued rising despite a broader slowdown, including on an RWA-focused platform built on Hyperliquid.
RWA deposits rise while DeFi liquidity contracts
The most notable signal in CoinShares and Token Terminal’s report is the divergence between RWA growth and the broader health of DeFi lending and staking liquidity. While total DeFi deposits fell by about 15% year over year, RWA deposits expanded sharply—more than tripling to $7.4 billion in Q2 2026.
This pattern matters because it suggests that RWA onboarding is not simply an extension of generalized risk appetite across crypto. If investors were responding mainly to a favorable overall market cycle, RWA and non-RWA flows would likely track each other more closely. Instead, the report frames RWA growth as being driven by “financial utility” such as collateral value, yield generation, and tradable access to real-world exposure.
CoinShares’ CEO characterized the dynamic as an asset class expanding through a downturn in its host ecosystem—an interpretation consistent with the report’s broader conclusion that RWAs are transitioning into a more durable role inside onchain markets.
Why yield-bearing tokens and Treasuries are winning
Within RWAs, the report identifies yield-bearing stablecoins and tokenized Treasury products as the dominant categories showing meaningful onchain usage. In Q2, Sky Protocol’s sUSDS led the category, giving holders exposure to a yield-generating version of its USDS stablecoin.
Tokenized Treasury funds are also highlighted as a major source of onchain collateral. The report points to BlackRock’s USD Institutional Digital Liquidity Fund (BUIDL) among the products being used in decentralized lending markets, where yield-generating assets can play a role in collateral strategies.
CoinShares and Token Terminal further suggest that the RWA landscape currently spans a broad yield spectrum. The report states that RWA products offer yields ranging from about 3.2% to 5.5%, with Treasury-oriented products toward the lower end of that range and higher-yield strategies carrying additional risks.
For investors and DeFi participants, the key takeaway is structural: tokenized Treasuries and yield-bearing stablecoins are not just “wrapped exposure,” but assets designed to plug into DeFi systems where liquidity, collateral management, and yield accounting can be operationally valuable—particularly when broader deposit trends are weakening.
Gold tokens and yield dollars power DEX activity
The report also describes where RWAs show up most clearly in day-to-day trading. On decentralized exchanges, gold-backed tokens and yield-bearing dollar products account for a large share of RWA trading activity.
CoinShares classifies gold-backed stablecoins such as Tether Gold (XAUt) and Paxos Gold (PAXG) as tokenized gold within its RWA reporting framework. The report indicates that these products generate significant trading volume as market participants trade around gold price swings, effectively turning gold exposure into an onchain, secondary-market instrument.
On the dollar side, yield-bearing products such as Ethena’s sUSDe are also described as contributing to RWA spot activity. While RWAs are often discussed in terms of long-horizon investment access, the report emphasizes how trading behavior on DEXs increasingly resembles active market participation—an important distinction from one-time minting and holding.
Supporting that view, the report notes that RWA spot trading volumes rose roughly 220% year over year. This occurred even while overall DEX volumes fell by about 70%, reinforcing the idea that RWAs are gaining traction as tradable assets with a growing secondary-market role, rather than riding broader DEX tailwinds.
RWAs are moving into leveraged derivatives
Beyond spot trading and lending collateral, the report highlights an expansion of RWA exposure into derivatives—particularly perpetual futures, where traders can take leveraged positions without owning the underlying tokenized asset.
RWA perpetual futures trading, the report says, has continued growing despite a broader slowdown in crypto-native derivatives markets. It points to an RWA-focused perpetual futures venue on tradeXYZ, built on Hyperliquid, where trading volume has increased by roughly 20 times since launch.
The report adds that activity has concentrated around commodities and equity index exposure, citing products tied to the S&P 500 and Nasdaq-100, along with technology stocks. Open interest has also been rising, suggesting that participation is not limited to brief bursts of speculation.
For traders and market makers, this matters because derivatives liquidity can change how RWAs are priced and hedged across onchain markets. As leveraged instruments become more common, RWAs may attract a wider range of participants—though it also introduces additional risks typical of leverage and derivatives exposure.
What to watch next
With RWA deposits accelerating while broader DeFi deposits contract, the next question is whether this pattern persists beyond Q2 2026 and whether growth remains concentrated in yield-bearing stablecoins and tokenized Treasuries—or broadens further into other real-world categories. Traders and DeFi users should also watch how expanding derivatives access affects liquidity, hedging behavior, and the risk profile of onchain RWA strategies.
Aave, the largest DeFi lending protocol with $14.3 billion in deposits, is moving to deprecate 50 low-adoption asset reserves and fully wind down its deployments on Sonic, Scroll, zkSync, Metis, Soneium and Aptos, under a governance proposal posted Wednesday by risk provider LlamaRisk. The changes… Read the full story at The Defiant
OpenAI has acquired patent assets from AI chip startup Rain AI after earlier negotiations to buy the company ended without an agreement, according to The Information.
Summary
- OpenAI acquired some Rain AI patents, but the financial terms and number of assets were not disclosed.
- Earlier negotiations for a complete takeover failed to produce a deal, leaving Rain AI close to shutting down.
- Rain AI developed energy-efficient chips for AI workloads and previously secured backing from Sam Altman.
- The transaction expands OpenAI’s hardware assets as advanced computing capacity becomes a strategic priority.
OpenAI buys Rain AI patents after failed takeover
OpenAI purchased some of Rain AI’s patents after the companies were unable to complete a broader acquisition, The Information reported. Neither company has publicly disclosed the price, the number of patents transferred, or the technologies covered by the transaction.
Rain AI has nearly ceased operating following the failed takeover talks, according to the report. Most of its employees have also left the startup, although it remains unclear whether any former Rain AI staff joined OpenAI as part of a separate arrangement.
The deal allows OpenAI to obtain selected intellectual property without assuming ownership of the entire company. It may also give the ChatGPT developer access to chip designs or related research that could support its work on AI computing infrastructure.
However, OpenAI has not explained how it plans to use the patents. Any connection between the acquired assets and a future OpenAI chip remains unconfirmed.
Rain AI pursued energy-efficient AI chips
Founded around eight years ago, Rain AI worked on specialized processors designed for artificial intelligence workloads. Its technology focused on lowering the energy and computing costs of training and running AI models.
Rain developed what it called neuromorphic processing units, or NPUs. The company sought to build chips that reproduced some features of the human brain while supporting both AI training and inference.
Rain’s investors included Altman, who made a personal investment in the company. The relationship later drew attention because OpenAI also explored commercial arrangements with the startup.
In 2019, OpenAI signed a nonbinding letter of intent to purchase $51 million worth of Rain AI chips once they became available, according to documents previously reviewed by Wired. The agreement did not result in a completed chip purchase.
Rain later struggled to secure sufficient funding and commercial demand. Reports in 2025 said the company began exploring a sale after a planned $150 million Series B funding round fell through.
Patent deal adds to OpenAI’s hardware push
The Rain AI transaction comes as OpenAI seeks more control over the infrastructure used to develop and operate its models. Advanced AI systems require large numbers of accelerators, considerable electricity and extensive data center capacity.
Purchasing patent assets could help OpenAI bring additional chip expertise in-house or strengthen its position when working with external semiconductor companies. It could also prevent potentially useful technology from disappearing as Rain AI winds down.
The arrangement does not establish that companies backed personally by Altman automatically receive financial support from OpenAI. Instead, the limited patent purchase suggests OpenAI selected specific technological assets after deciding against acquiring Rain AI outright.
Altman’s investment makes the transaction relevant from a governance perspective, although the available reporting does not identify any conflict or describe his involvement in approving the deal.
US oversight expands alongside AI infrastructure
OpenAI’s hardware expansion is unfolding as Washington increases its focus on advanced AI systems and the infrastructure supporting them.
Crypto.news reported in late July that OpenAI and Anthropic were working with the Trump administration on a consistent federal review process for models with advanced cybersecurity or national security capabilities. Despite competing for customers, researchers and computing resources, both companies sought standards that would apply across the industry.
Under the federal framework, developers could provide designated “covered frontier models” to the government for evaluation for up to 30 days before releasing them to other trusted partners. The White House finalized a voluntary framework in early August, though questions remain about which models will qualify.
The Rain AI patent acquisition addresses a different part of OpenAI’s strategy, but both developments reflect the growing importance of U.S.-controlled technology, computing capacity and intellectual property. OpenAI has not announced whether the Rain patents will support an internal chip project, a licensing program or future partnerships.
FBI Reportedly Returns to Search Nancy Guthrie’s Tucson Neighborhood as Case Nears Six Months
Morgan Stanley Launches Spot Ether, Solana ETPs on NYSE Arca
Blake Monroe makes surprising confession about her WWE entrance theme
-
Fashion6 days agoWeekend Open Thread: Wit & Wisdom
-
Politics6 days agoMeta enters AI-training agreement with far-right ‘propaganda rag’ Newsmax
-
Politics4 days agoZack Polanski: an incitement to murder Nigel Farage?
-
Crypto World6 days agoMicroStrategy Post-Earnings CLARITY Act Push Could Add New Catalyst for Its Stock
-
Crypto World6 days agoXRP Ledger v3.3.0 brings five institutional features
-
Crypto World6 days agoNew York sues Kalshi over prediction market gambling
-
Crypto World5 days agoCrypto PAC spending tops $2M in Michigan House race
-
Business4 days agoDTCR: Deleveraging And A Hedge Fund Collapse Point To A Possible AI Bottom
-
Business7 days agoTrump Announces Hamas Disarmament Agreement as Iran Strikes Kuwait Air Base and US Attacks Pause Overnight
-
Crypto World6 days ago3 Fed Officials Just Explained Their Rate Hike Vote: Is Inflation Winning?
-
Tech5 days agoESET tracks rise in malicious AI skills and adaptable malware
-
Crypto World4 days ago
XRP Ledger urges node upgrade after manifest flood
-
Tech7 days agoGemini Spark can now use Chrome logins and saved passwords to run errands on your behalf
-
Tech7 days agoBuilding A Reproduction PlayStation Motherboard
-
Sports6 days agoFrance Cricket implodes: letters hidden in a drawer and a board at war
-
Crypto World6 days agoMoneyflip CEO charged in $40K murder-for-hire plot
-
Sports6 days agoBruno Fernandes decision made as Man United ‘discuss’ striker transfer option
-
News Videos5 days agoFinancial Crash Expert: The 90-Day Collapse Timeline They Are Desperately Hiding.
-
NewsBeat7 days agoSpain sending troops as thousands enter enclave of Ceuta from Morocco
-
Sports6 days agoCommonwealth Games 2026 Live Updates | Day 9 CWG 2026: Lovlina Borgohain, Sachin Siwach Enter Final After Indian Judokas Script History
You must be logged in to post a comment Login