There are now 160+ stablecoins, and almost nobody is talking about how they trade against each other. USDS, sUSDS and the Sky Savings Rate are quietly sitting underneath the answer.
160 stablecoins. 12,720 possible pairs. Or one shared base pair.
Here is a number almost nobody in crypto has looked at.
Every day, $9.6 trillion moves through the global foreign exchange market. The US dollar sits on one side of 89.2% of those trades, according to the BIS Triennial Central Bank Survey.
Not because the dollar is the best currency in the world. Because it is the one everybody already holds.
Stablecoins just ran the same experiment. Much faster.
In four months, a stablecoin FX layer processed more than $12 billion in cumulative volume.
In its first full quarter it handled $14.9 billion. It went live on 25 June 2026 with roughly $150 million in seed liquidity.
The asset sitting at the center of it is USDS.
What Is a Stablecoin FX Layer, and Why Does One Exist?
A stablecoin FX layer is shared liquidity infrastructure that lets different stablecoins trade against each other through one common base pair, instead of each one funding its own isolated pools.
That sounds boring. It is not.
Think about how traditional FX actually works. You do not trade Thai baht directly against Chilean peso.
You route through the dollar. One deep market quietly does the work of thousands of thin ones.
Stablecoins never got that layer. Every new issuer has had to bootstrap liquidity from scratch, against every asset its users already hold. PayPal did it.
Ripple did it. The next hundred banks and fintechs were lining up to do it again.
Spark, an independent capital allocator in the Sky Agent Network, launched its Stablecoin FX Layer on Uniswap v4 to solve exactly that. The first two markets were USDS/USDT and USDS/PYUSD.
Sam MacPherson, CEO of Spark, framed the problem plainly. The market does not need hundreds of shallow new pools.
It needs a way to concentrate liquidity so the same capital can work across hundreds of stablecoins.
You do not trade Thai baht directly against Chilean peso. You route through the dollar. Stablecoins never got that layer.
Why 160 Stablecoins Create 12,720 Separate Problems
Here is the math that turns stablecoin liquidity fragmentation from an inconvenience into a structural issue.
DefiLlama tracked more than 160 stablecoins above $1 million in market cap in November 2026. If every one of them needs a direct market against every other one, you get n(n-1)/2 pairs.
That is 12,720 markets. Each needing its own capital. Each sitting half empty.
Route everything through one shared base pair instead, and that number drops to 160.
What changes when you do that:
Capital gets concentrated rather than duplicated across thin poolsA new issuer plugs into existing depth instead of starting from zeroSpreads tighten, because depth is pooled rather than scatteredLiquidity starts behaving like infrastructure rather than a private moatIsolated pairs scale as n(n-1)/2. A shared base pair scales as n. That gap is the whole argument.
Now layer in the market backdrop. Total stablecoin supply was about $288.87 billion on 16 November 2026, down from a May peak near $309.7 billion.
The pie is not growing right now. The number of slices is.
That is the trap. More issuers, flat supply, thinner individual pools, worse execution for everyone.
Four Months, $12 Billion, Then $14.9 Billion in a Single Quarter
The Stablecoin FX Layer went live on 25 June 2026 with roughly $150 million migrated into two Uniswap v4 pools. Coverage at the time called it one of the largest AMM liquidity migrations in DeFi history.
By 23 September 2026, cumulative volume had passed $12 billion. Spark’s Q3 2026 financial report then put the quarter’s figure at $14.9 billion, describing the layer as a core venue for stablecoin conversion on Ethereum.
Ripple’s RLUSD shows how quickly a new issuer can scale on shared rails:
RLUSD joined in September 2026 with a $375 million deployment$50 million of that went to exchange liquidity, roughly $25 million RLUSD paired with $25 million USDSThe RLUSD/USDS market cleared more than $600 million in week oneIt crossed $1 billion in cumulative volume in under a month, averaging around $40 million a dayBy late October, the layer was funding both sides of that pool, so the issuer did not have to commit the capital itself$150 million of seed liquidity turned into $14.9 billion of quarterly volume in one quarter.
Hayden Adams of Uniswap Labs noted that Uniswap handled close to $400 billion in stablecoin swaps during 2026, and called the first three months of the collaboration a strong start.
Why USDS Became the Base Pair Everything Routes Through
Every FX market eventually picks a vehicle currency. The winner is never the most sophisticated asset. It is the one with the deepest, most boring, most available liquidity.
USDS got picked for three practical reasons.
It converts at 1:1 with no slippage. Sky Protocol’s Peg Stability Module lets USDC and USDT convert directly into USDS at a fixed rate, with zero fees and no market impact. For a routing asset, predictable entry and exit matters more than anything else.
Its backing is checkable in real time. Protocol Collateral stood at $12.80 billion against 8.55 billion of stablecoin supply on 17 November 2026. That is published onchain, not attested quarterly.
It was already deep. USDS-supported pairs accounted for more than 50% of Ethereum stable-to-stable volume on Uniswap during August 2026. The liquidity was there before the layer needed it.
Traditional FX settled on the dollar. Onchain FX is settling on a base pair of its own.
One more detail worth sitting with. A larger portion of the $3.69 billion of USDS drawn through Spark now supports liquidity inside the FX Layer, according to Sky Protocol’s Q3 2026 reporting.
That is a capital allocation network quietly becoming exchange infrastructure.
Where the Sky Savings Rate and sUSDS Actually Fit
This is the part most coverage skips, and it is the most interesting mechanic in the whole design.
Traditional market making demands that capital sit still. You commit it to a pool and it waits there, whether or not anybody trades that day.
The FX Layer is being built differently. Liquidity is made available dynamically for swaps, while capital not currently required for exchange can remain productive.
That is where the Sky Savings Rate comes in.
The Sky Savings Rate is a variable rate set by SKY-token-holder governance. It is funded by revenue that Sky Protocol accrues as independent allocators, including Spark, Grove, Keel, Obex and Osero, draw USDS liquidity and deploy it into their own strategies.
sUSDS is how that rate is accessed. It is the world’s largest yield-generating stablecoin, it compounds programmatically, and it has no lockups.
Where things stood on 17 November 2026:
Sky Savings Rate: 3.80% APY, raised by governance on 7 October 2026Protocol Collateral backing USDS: $12.80 billionSavings supply held in sUSDS: 3.78 billionNet Protocol Revenue for Q3 2026: $37.19 million, a sixth consecutive positive quarterAnnualized Gross Protocol Revenue run-rate: $397.2 millionThe collateral, supply and rate figures sitting underneath the base pair.
So the exchange layer and the savings layer are not two separate products bolted together. Idle exchange capital has somewhere to go. That is the point.
Who Is Actually Trading on It?
I assumed arbitrage bots. The data says otherwise.
Between 20 September and 19 October 2026, the RLUSD/USDS pool processed about $453 million across 2,410 transactions.
Roughly $167 million tied to yield and lending activityAbout $130 million moving through exchange-wallet routesAround 30.5% unclassifiedSpark’s own liquidity management: about 4% of the totalNearly two thirds of pool volume came from yield, lending and exchange wallets, not from the venue operator.
Nearly two thirds of it is real financial plumbing. Lending protocols rebalancing. Exchanges moving inventory. Treasuries shifting between dollar assets.
That 4% figure matters most. When the operator of a venue accounts for almost none of its volume, the activity is organic.
What Happens When the Next Hundred Stablecoins Launch
Bank-issued stablecoin supply went from $337 million in May 2026 to $1.08 billion by 16 November 2026, led by SoFi, Zand Bank and Societe Generale. Citi has projected the stablecoin market could reach $4 trillion by 2030.
Every one of those issuers hits the same first question. Where does liquidity come from?
The old answer was: fund it yourself, pair by pair, and hope.
The new answer is: plug into a base pair that already has depth behind it.
If that model holds, the competitive ground shifts underneath the whole sector. Issuers compete on distribution, compliance and user experience. Liquidity stops being a moat and starts being shared infrastructure.
Galaxy added $100 million of sUSDS to its corporate treasury in September 2026 and approved it as collateral across its institutional trading business. That is what the demand side of this looks like when it shows up.
The Open Question I Keep Coming Back To
$12 billion in four months is not a victory lap. It is an early signal.
The honest open question is whether a shared base pair concentrates risk as much as it concentrates liquidity.
If most stable-to-stable routing depends on one asset, that asset’s collateral quality becomes everybody’s problem.
Transparency helps. Every position is onchain and independently checkable, which is more than can be said for most of traditional FX.
But I would rather hear your read than give you mine.
Does a shared base pair make the stablecoin market safer, or does it just move concentration risk somewhere harder to see? Drop it in the comments.
The Stablecoin FX Layer Did $12 Billion in Four Months was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story.
