
DeFi Lending Still Pays Near 5% on Stablecoins as Yields Hold Above the Fed
Lending dollars in decentralised finance still pays about 4.9 percent, roughly a point more than the United States Federal Reserve’s benchmark rate, according to a snapshot of DeFi lending markets from November 5, 2026. The figures, drawn from DeFiLlama yields data and refreshed hourly, show a market in which the headline return on supplying stablecoins has held up even as traditional rates have drifted lower, and in which the gap between what a lender actually earns and what a headline annual percentage yield suggests can be very large. The average is weighted by the dollars supplied to each pool, so it reflects where the money really sits rather than the highest rate on the smallest pool. USDT pays about 4.0 percent on that basis, and ETH pays about 1.3 percent
Those billions are concentrated in a small number of very large pools. The largest USDC pool is Maple’s Syrup USDC, with 2.5 billion dollars supplied at 5.10 percent, all of it base interest. Aave’s Ethereum USDT pool holds 2.6 billion dollars at 3.74 percent, and Aave’s Ethereum USDC pool holds 2.3 billion dollars at 4.52 percent. Spark’s USDS pool holds 1.4 billion dollars at 2.80 percent. Together, those four pools alone account for more than 8 billion dollars of supplied capital. The market is dominated by large, plain, mostly base-interest pools.
DeFi Lending Rates: The Biggest Pools and the Highest Yields
The full table of lending rates by pool, sorted by dollars supplied, tells a more detailed story than the headline average. Maple’s Syrup USDC leads the large pools at 5.10 percent on 2.5 billion dollars, with every point of that return coming from base interest paid by borrowers rather than from token rewards. Aave’s Ethereum USDC pool pays 4.52 percent on 2.3 billion dollars, also entirely base interest. Aave’s Ethereum USDT pool pays 3.74 percent on 2.6 billion dollars, the single largest pool in the table. Spark’s USDS pool pays 2.80 percent on 1.4 billion dollars.
Move away from the giants and the rates climb, but so do the caveats. Jupiter Lend’s Solana USDC pool pays 5.76 percent on 465 million dollars, all base interest Morpho’s USDe pools on Base and on Robinhood Chain each pay 5.25 percent, but every point of that return is token rewards, with zero base interest, which is a very different proposition from Maple’s 5.10 percent of borrower-paid interest. Aave’s Base USDC pool pays 6.05 percent on 199 million dollars, split between 3.98 percent base interest and 2.07 percent rewards. Compound’s Ethereum USDT pool pays 6.47 percent on 158 million dollars, all base interest.
Base Interest vs Token Rewards: Reading the APY Honestly
Every rate in the table is shown as an annual percentage yield, defined as base interest from borrowers plus token rewards, and the split between the two is the single most important detail for anyone comparing pools. Base interest is money that borrowers pay for the use of the supplied asset, and it is the durable part of the return. It rises and falls with utilisation, but it is generated by real demand for the asset. Token rewards are extra tokens that a protocol hands out as an incentive, and they can end without warning or lose value if the reward token’s price falls. A pool paying 5 percent entirely in base interest and a pool paying 5 percent entirely in rewards may show the same headline number, but they are not the same investment.
There is also a timing dimension that the headline yield hides. Rates on the page are refreshed hourly from DeFiLlama data, and the snapshot used here was taken at 9:29 PM Eastern Time on November 5, 2026. A rate observed at that moment can move within hours as utilisation changes, as a large borrower repays or a large supplier withdraws, or as a reward programme starts or ends. The page is explicit that its rates change on each venue’s own schedule and that the figures exclude trading fees, spreads, gas costs, and liquidation risk, all of which eat into the realised return. Read the table as a photograph of a moving market, not a menu of guaranteed returns.
Pooled vs Isolated Markets: How Aave, Morpho and the Others Differ
Behind the rates sits a structural choice that matters as much as the numbers: how each protocol organises its markets. Aave uses pooled liquidity markets, where suppliers deposit supported assets into a shared pool, borrowers post collateral, and rates respond to market utilisation. Parameters differ by asset and deployment, so there is no single Aave-wide rate or liquidation threshold, and users should check the selected market rather than assuming one figure applies everywhere.
Morpho takes a different approach, offering permissionless isolated lending markets and vaults that can allocate deposits across selected markets. In an isolated market, each collateral-and-loan pair stands on its own, so a problem in one pair does not automatically contaminate the others. For vaults, the important detail is the curator’s role, the limits they set, and the allocation choices they make, because the depositor is trusting the curator’s judgement as well as the code. This is different from Morpho’s earlier positioning as a peer-to-peer layer over Aave and Compound, and it reflects a broader shift in DeFi toward curated vaults in which a named manager decides where deposits go.
Compound organises each market around a base asset, and supplying that base asset can earn interest while assets supplied as collateral support borrowing but do not automatically earn interest. Spark and Sky provide overcollateralised lending and savings products connected to the Sky ecosystem, which centres on USDS and related savings and borrowing tools, Maple combines onchain infrastructure with managed lending and credit strategies, so returns can depend on borrower repayment and strategy management rather than only on an automated overcollateralised pool. Kamino provides lending, borrowing, and vault products on Solana, The practical lesson is that the protocol’s name alone tells a lender very little. The market structure, the curator, the collateral rules, and the withdrawal terms are what determine the risk.
The Risks Behind the 5% Yield
A return of about 4.9 percent on dollars looks generous next to a Fed funds midpoint of 3.875 percent, but the gap exists because the risks are real and the deposits are not insured. The page’s own risk section lists them plainly. Smart-contract risk means a bug, an exploit, or a faulty upgrade can cause losses or freeze access, and audits and bug bounties can reduce that risk without guaranteeing safety. Liquidation and oracle risk means borrowers can lose collateral when a position crosses its liquidation threshold, and bad or delayed price data can worsen outcomes. Token and stablecoin risk means a supplied asset, a collateral token, or a stablecoin can lose value or its peg Liquidity risk means high utilisation or a withdrawal queue can delay access to funds.
Credit and counterparty risk applies most to managed or undercollateralised strategies, which depend directly on borrowers, managers, and service providers rather than only on overcollateralised smart contracts.
The history of the sector is a warning against treating any of this as theoretical. The 2022 crypto credit crisis put Celsius, BlockFi, Voyager, and Genesis lending entities into bankruptcy, and many customers became creditors rather than depositors. The lesson carries across the divide: a high yield is a price for risk, not a free lunch. The page’s own checklist is the right way to approach it: verify the official protocol address, network, and token contract, separate base returns from temporary rewards, check liquidity, utilisation, supply caps, and withdrawal mechanics, read the risk parameters, audits, incident history, and admin controls, and start with an amount that can be afforded to be lost, testing a small withdrawal before committing more.
One figure in the table deserves a final note, because it explains a puzzle that confuses many newcomers. Lending ETH pays only about 1.3 percent, far below the return on stablecoins, and the reason is straightforward: most ETH in lending pools is posted as collateral rather than borrowed, so there is little demand pushing its rate up. The market pays for what borrowers want, and borrowers want dollars, not more ETH. That single fact explains the shape of the whole table, and it is a useful reminder that DeFi lending rates are not set by generosity. They are set by demand, and the demand is for stablecoins at a modest premium over the Fed.
FAQ
How much does lending dollars in DeFi pay right now?
About 4.9 percent on a weighted-average basis as of the November 5, 2026 snapshot, roughly a point above the Fed funds midpoint of about 3.875 percent. USDT pays about 4.0 percent and ETH about 1.3 percent.
Which are the largest stablecoin lending pools?
Aave’s Ethereum USDT pool holds 2.6 billion dollars at 3.74 percent, Maple’s Syrup USDC holds 2.5 billion dollars at 5.10 percent, Aave’s Ethereum USDC holds 2.3 billion dollars at 4.52 percent, and Spark’s USDS holds 1.4 billion dollars at 2.80 percent.
What is the difference between base interest and token rewards?
Base interest is paid by borrowers and is the durable part of the return. Token rewards are extra tokens a protocol hands out, which can end or lose value. Morpho’s USDe pools pay 5.25 percent entirely in rewards, while Maple’s Syrup USDC pays 5.10 percent entirely in base interest.
What is the difference between pooled and isolated lending markets?
Aave uses pooled markets where many suppliers and borrowers share one pool, while Morpho offers isolated markets where each collateral-and-loan pair stands alone, plus vaults whose curators allocate across selected markets. Compound organises markets around a base asset, and Maple uses managed credit strategies.
What are the main risks of DeFi lending?
Smart-contract bugs, liquidation and oracle failures, stablecoin depegs, withdrawal delays at high utilisation, credit and counterparty risk in managed strategies, rewards ending or losing value, governance and admin-key changes, and wallet risks such as phishing. Deposits are not insured.
Source: DeFi Rate (DeFiLlama yields data)
Related reading on MM Crypto Analysis Web
- Lido Adds Firelight Loss Protection to EarnETH and EarnUSD Vaults
- Polygon Hard Fork Proposal Would Return 70.49 Million Frozen POL to Binance and Polymarket
- Pendle Trading Volume Hits a 30-Day High as Yield Demand Returns
More in DeFi on MM Crypto Analysis Web.