
Aave · Best for: The safest large lending market in DeFi
The reference lending protocol: no depositor has lost funds to a protocol failure in eight years of operation.
Topics
Latest
Lending markets ranked on collateral policy, oracle design and loss record. Every score below is the weighted mean of the rubric marks, not an editor's gut feel.
| # | Service | Score | Best for | Fees | Since |
|---|---|---|---|---|---|
| 1 | 9.0/10 | The safest large lending market in DeFi | |||
| 2 | 8.4/10 | Minimal, immutable lending primitives | |||
| 3 | 8.4/10 | Cheap DAI and USDS borrowing | |||
| 4 | 8.3/10 | Borrowing against blue-chip collateral at scale | |||
| 5 | 8.1/10 | Capital-efficient borrowing with combined liquidity | |||
| 6 | 8.0/10 | Isolated markets for long-tail collateral | |||
| 7 | 7.9/10 | Conservative single-borrow-asset markets | |||
| 8 | 7.9/10 | Solana lending with proper risk parameters | |||
| 9 | 7.6/10 | Customisable lending vaults for any asset | |||
| 10 | 7.4/10 | Soft liquidations that do not wipe the position | |||
| 11 | 7.3/10 | Avalanche lending | |||
| 12 | 7.1/10 | Solana lending with cross-margin | |||
| 13 | 6.6/10 | BNB Chain lending liquidity | |||
| 14 | 6.2/10 | Tron ecosystem lending and energy rental | |||
| 15 | 4.9/10 | Cross-chain lending, with serious caveats |

The reference lending protocol: no depositor has lost funds to a protocol failure in eight years of operation.

Immutable lending markets with the risk decisions pushed out to named curators — better rates, and someone specific to blame.

An Aave fork with the Sky balance sheet behind it, offering the cheapest stablecoin borrowing available.

The oldest credit system in DeFi, still solvent after Black Thursday, and increasingly dependent on assets that live off-chain.

Instadapp's lending layer merges DEX and lending liquidity for unusually high loan-to-value ratios — clever, and newer than its competitors.

Built entirely around isolation, so a bad listing can only ever hurt the people who chose it.

The protocol that started algorithmic lending, now a conservative and much smaller alternative to Aave.

The most carefully parameterised lending market on Solana, and the one that handled the chain's volatility without bad debt.

Rebuilt from the ground up after a $197m exploit, with the returned funds and the rewritten architecture both on the record.

The only lending design that converts your collateral gradually instead of liquidating it — genuinely novel, and still proving itself.

Avalanche's main money market, competently run on a chain with less and less to lend against.

A capable Solana lending market whose 2024 leadership walkout showed how much of DeFi still depends on a few individuals.

The dominant BNB Chain lending market, with a bad-debt history that came directly from listing collateral it should not have.

Large by deposits, opaque by governance, and inseparable from the Tron ecosystem's central figure.

Exploited twice, most recently for $50m through compromised multisig signers — the clearest case in this list of keys mattering more than code.
Score = Σ (mark × weight), rounded to 0.1. A criterion with nothing published scores 0.
Every lending protocol failure in this sector traces back to one of two things: collateral that should not have been accepted, or a price feed that could be pushed. Those two criteria therefore carry nearly half the weight in this rubric, ahead of rates, which are the thing most users compare first.
Loss record is scored on conduct as much as size. A protocol that took bad debt and recapitalised through governance is in a different position from one that quietly left lenders short. Where a protocol has been exploited, we look at whether funds were recovered, whether the fix was structural, and whether the same class of bug remains possible.
A supply rate far above the market usually means one of three things: emissions subsidising it, borrowers who cannot get credit elsewhere, or collateral nobody else would accept. We check which before scoring the rate favourably.
Isolated markets and per-asset caps mean a single bad listing no longer threatens every depositor. Protocols that have adopted that architecture score better on collateral policy than those still running one shared pool for every asset.