What stablecoins earn, and what you take on to earn it
The same dollar earns anything from 1% to 8% depending on what you are willing to accept. These are grouped by that, not by the size of the number.
Rates in different groups are not comparable, and the groups are ordered by how clearly you can see the way the money is lost — not by how much they pay. Reading down the page, the rates rise because the risk becomes harder to see, not because anyone is being more generous.
The bold number is the 30-day average, which is roughly what you would have earned. Today’s rate sits next to it in grey and is often well off it. Incentives is the share of the rate paid in a protocol’s own token, which is a budget that can be cut at any time.
Lending against collateral worth more than the loan
Borrowers post more value than they take out, and the protocol sells that collateral automatically if it falls too close to the debt. What protects a lender here is that margin, not anybody’s promise to repay.
What goes wrong: A bug in the contracts, a price feed reporting the wrong number, or a crash fast enough that collateral sells for less than the debt and the shortfall lands on lenders. Rates float with how badly people want to borrow, so a good week is not a rate you keep.
| Protocol | Asset | 30-day avg | Now | Incentives | Size | Track record |
|---|---|---|---|---|---|---|
| Sky | SUSDS Ethereum | 3.55% | 3.60% | none | $4.6B | 1 year+ |
| Aave v3 | USDE Ethereum | 3.07% | 4.75% | 84% | $527M | 2.3 years |
| Aave v3 | USDT Ethereum | 3.35% | 3.16% | none | $426M | 3.6 years |
| Sky | SUSDS Arbitrum One | 3.55% | 3.60% | none | $363M | 212 days |
| Aave v3 | USDC Ethereum | 3.87% | 3.51% | none | $207M | 3.6 years |
| Sky | SDAI Ethereum | 1.25% | 1.25% | none | $203M | 3.8 years |
| Aave v3 | SGHO Ethereum | 4.35% | 4.50% | none | $177M | 120 days |
| Fluid | USDC Ethereum | 4.89% | 4.49% | none | $154M | 2.3 years |
Vaults where someone else chooses the markets
You deposit and a curator — a firm paid to make the call — spreads the money across lending markets on your behalf. The underlying loans are the overcollateralised kind above.
What goes wrong: Everything in the row above, plus the curator’s judgement. They can move into a market you would not have chosen, and the rate tells you nothing about who they are. Find out which firm curates a vault before the number matters.
| Protocol | Asset | 30-day avg | Now | Incentives | Size | Track record |
|---|---|---|---|---|---|---|
| Morpho | STEAKUSDC Base | 4.26% | 4.33% | none | $429M | 285 days |
| Morpho | GTUSDCP Base | 4.27% | 4.33% | none | $420M | 285 days |
| Morpho | SIRLOINUSDC Base | 5.67% | 5.61% | 49% | $399M | 104 days |
| Morpho | SENRLUSDV2 Ethereum | 6.28% | 5.78% | 57% | $393M | 192 days |
| Morpho | SENPYUSDMAIN Ethereum | 5.78% | 5.43% | 58% | $383M | 264 days |
| Morpho | USDE Base | 4.64% | 4.75% | 100% | $352M | 95 days |
| Morpho | SPARKUSDC Base | 3.83% | 3.88% | none | $278M | 285 days |
| Morpho | SENPYUSDPRIMEV2 Ethereum | 6.21% | 5.95% | 42% | $197M | 124 days |
Dollars that earn from a trading position
The token holds crypto together with an offsetting short position, which keeps its value near a dollar. The yield is what leveraged traders pay to keep their positions open.
What goes wrong: That payment can turn negative, and has. The yield then stops or reverses. It also depends on the exchanges holding the collateral staying solvent and accessible, which is a different kind of risk from anything on-chain.
| Protocol | Asset | 30-day avg | Now | Incentives | Size | Track record |
|---|---|---|---|---|---|---|
| Ethena 7 days unstaking | SUSDE Ethereum | 4.56% | 4.91% | none | $1.3B | 2.6 years |
A claim on something off-chain
Treasury bills, money-market funds and credit funds, wrapped as a token. The rate mostly comes from ordinary interest rates rather than from crypto.
What goes wrong: The rate is the least of it. An issuer and a custodian stand between you and the actual asset, and what you own is what the legal wrapper says you own. Redemption can be slower than holding a token suggests, and in a stressed market it can be suspended.
| Protocol | Asset | 30-day avg | Now | Incentives | Size | Track record |
|---|---|---|---|---|---|---|
| Centrifuge Janus Henderson Treasury Fund | USDS Ethereum | 3.40% | 2.88% | none | $725M | 116 days |
| Centrifuge Janus Henderson AAA CLO Fund | USDC Avalanche C-Chain | 5.10% | 4.99% | none | $262M | 116 days |
| Bitwise | USCC Ethereum | 5.39% | 7.65% | none | $77M | 340 days |
| Midas mTBILL | USDC Ethereum | 3.29% | 3.33% | none | $73M | 102 days |
| Midas mFONE | USDC Ethereum | 8.18% | 7.67% | none | $65M | 102 days |
| Centrifuge Janus Henderson AAA CLO Fund | USDC Base | 5.10% | 4.99% | none | $51M | 97 days |
| Midas mHYPER | USDC Ethereum | 6.37% | 7.12% | none | $32M | 102 days |
Lending to firms that have posted no collateral
Money is lent to trading firms and businesses against their creditworthiness. Where a vault is named after a firm, that firm is the borrower — the name is the risk.
What goes wrong: If a borrower does not repay there is nothing to seize and nothing to liquidate; the loss is shared across everyone in the pool. Withdrawals can also queue, because the money is out on loan and cannot be recalled on demand. The extra percent or two is the payment for accepting exactly that.
16,962 pools scanned, 28 shown. Data from DefiLlama, 13 September 2026. Each protocol name links to that pool’s page on DefiLlama so you can check these figures against the source rather than taking them from us.
What is left out, and why
A pool appears here only if it holds at least $10M, has at least 90 days of history, pays something for simply being supplied, and comes from a protocol we are prepared to describe in the paragraphs above. That last rule removes far more than the others: a risk note we cannot write is a risk we do not understand well enough to put a number next to.
The history rule matters more than it looks. Scanning the same data with it switched off puts a protocol eight days old at the top of the table, above every established one, purely because a new pool can advertise whatever it likes before anything has been tested. A table sorted by rate would have led with it.
Getting funds to the right chain
Most of the size here sits on Ethereum, and most of the cheap transacting does not. If a rate you want is on a chain your funds are not on, that is a bridge transfer — check what the crossing costs first, because on a small balance it can be worth more than a year of the difference in rate.
Common questions
Why is this not sorted with the highest rate first?
Because that ordering answers a question nobody should be asking. The rates on this page are not competing offers for the same thing — 7% for lending to a trading firm with no collateral and 4% for lending against collateral worth more than the loan are different products, and putting them in one ranked column implies the top one is simply better. Inside each group the order is by size, which is a rough measure of how much scrutiny a pool has already had.
Which number should I actually look at?
The 30-day average. Today’s rate moves with borrowing demand and can be double or half the average for a week at a time — one pool here currently advertises 7.65% against a 30-day average of 5.39%. The average is closer to what you would have earned.
What does the incentives column mean?
The share of the rate that is paid in a protocol’s own token rather than in interest from borrowers. Incentives are a marketing budget: they can be cut at any time, and the token can fall while you hold it. A pool paying 4.75% where all of it is incentives is a different proposition from one paying 4.75% from borrower interest.
Does a long track record mean a protocol is safe?
No, but a short one is genuinely informative. A protocol that has run for three years has been through market conditions a three-month-old one has not seen yet. Nothing below 90 days appears on this page at all, and 90 days is still short.
Where does this data come from?
DefiLlama’s public yields API, which tracks around 17,000 pools. This page keeps a few dozen: stablecoin pools, on the ten chains this site covers, from protocols we are willing to describe, holding at least $10 million and with at least 90 days of history. Every one of those rules is stated here rather than applied quietly.
Is this a recommendation?
No. Nothing here is advice, nothing is ranked as better, and there is no way to deposit anything from this page — it exists to show what the rates are and what each one costs you in risk. Read the protocol’s own documentation and make your own decision.
Nothing on this page is financial advice, and nothing here is a recommendation to use any protocol listed. Bospora has no relationship with any of them, earns nothing from them, and provides no way to deposit into them. Rates are historical and do not predict anything. Protocols can and do lose money.