Ranked & Reviewed
Best Crypto Lending Platforms
Lending markets ranked on collateral terms, liquidation mechanics, rate transparency and contract security.
No money changes hands. This is a non-commercial project: no advertising, no sponsored placements, and no affiliate or referral links. Links to the services below earn us nothing. Every overall score is the plain average of the criteria shown on the page, so you can check the arithmetic yourself — see our rating methodology and how the site is funded.
Morpho publishes its permitted loan-to-value set as nine exact percentages, caps the liquidation incentive at 1.15 by formula, takes no protocol fee on liquidations, and releases under GPL-2.0 — the only recognised open-source licence among the major lending markets here.
Pros
- GPL-2.0, a recognised open-source licence, on both Morpho Blue and MetaMorpho
- Permitted LLTV values published exactly: 0, 38.5, 62.5, 77, 86, 91.5, 94.5, 96.5 and 98 per cent
- Liquidation incentive capped by published formula at a maximum of 1.15
- Documentation states the protocol takes no fee from liquidations — the entire incentive goes to the liquidator
- Market fee capped in code at 25 per cent
Cons
- Owner functions are protected only by onlyOwner; no timelock is documented
- Only three audit reports are published, fewer than most peers here
- Market parameters are chosen from an approved set rather than being immutable
Venus publishes three tiers of governance proposal with exact voting periods and execution delays, and releases under BSD-3-Clause. Its liquidation documentation, however, gives only worked examples rather than the protocol's actual parameters.
Pros
- Three proposal types with published durations: normal 24h vote plus 48h delay, fast-track 24h plus 6h, critical 6h plus 1h
- BSD-3-Clause on both the core protocol and isolated pools repositories
- Additional cross-network execution delays stated separately
- Audits named and dated: Certik, Quantstamp and FairyProof
Cons
- Liquidation figures in the documentation are explicitly a hypothetical example, not live parameters
- Reserve factor and per-asset collateral factors are not published
- An Access Control Manager lets some actions bypass voting and execute via a guardian multisig
Kamino publishes its protocol spread market by market — 11% to 20% of interest paid — states outright that there are no deposit, withdrawal or origination fees, and documents its liquidation penalty scale from 0.1% to a 10% maximum.
Pros
- Protocol spread published per market: 11% SOL, 15% USDC and others, 20% in JLP and Altcoins markets
- States plainly there are no deposit, withdrawal or origination fees
- Liquidation penalty published: 0.1% immediate, scaling to a 10% maximum
- Close factor of 10% per liquidation event published
- Five audit firms named with dates, plus a $1.5m bug bounty
Cons
- Business Source License 1.1 until 17 November 2027, not an open-source licence
- Governance, upgrade authority and any timelock are not stated
- Per-asset liquidation LTV values are not published
Euler publishes hard bounds on what governance may do — an interest fee that must sit between 10% and 100%, a protocol share capped at 50% regardless of what is requested — and states those limits are enforced permanently for non-upgradeable vaults.
Pros
- Interest fee constrained to a guaranteed 10% to 100% range, enforced by the protocol
- Protocol fee share capped at 50%: vaults ignore any higher value returned by config
- States that for non-upgradeable vaults, limits on governance are enforced permanently
- Vaults can be created with no governor at all
- Twelve audit reports from ten firms including OpenZeppelin, Spearbit, ChainSecurity and Certora
Cons
- Business Source License 1.1 with a change date of 24 April 2029, the second longest here
- Maximum liquidation discount defaults to zero and must be set by each vault creator
- No timelock duration is stated
- Liquidation terms vary per vault rather than being protocol-wide
Aave publishes the deepest audit archive in this comparison — 49 reports from fifteen firms — and states a governance timelock of either one or seven days depending on proposal type. Its licence file states in its own words that it is not an open source licence.
Pros
- 49 audit reports published from fifteen firms including OpenZeppelin, Trail of Bits, Certora and Sigma Prime
- Most recent audits dated March 2026, so the archive is current
- Governance timelock stated as either one day or seven days depending on proposal type
- Liquidation bonus mechanics documented, always above 100%
Cons
- The LICENSE file states outright: the Business Source License is not an Open Source license
- Per-asset loan-to-value, liquidation threshold and reserve factor values are not published in the documentation
- Risk Admin and Pool Admin roles can adjust parameters, and Stewards hold delegated authority over some
Silo sets each market's liquidation fee at deployment and states it cannot be changed afterwards — a stronger guarantee than any timelock. Its licence runs under BUSL until 2031, the longest delay in this comparison.
Pros
- Liquidation fee set per market by the deployer and cannot be changed afterwards
- Hard cap in code: maximum fee 50%, with separate limits for deployer, flashloan and liquidation fees
- Audited by Certora with formal verification, plus Enigma Dark, Spearbit Cantina, Sigma Prime and a Code4rena contest
- SiloDAO granted explicit rights to operate the protocol under the licence
Cons
- Business Source License 1.1 with a change date of 31 January 2031, the longest here
- Actual fee values are set at deployment and not published in the documentation
- Liquidation thresholds are referenced but their values are not stated
- Factory uses Ownable2Step with no timelock documented
Spark publishes the health-factor threshold at which a liquidator may take the whole position rather than half — at or below 0.95, the close factor rises from 50% to 100%. Its core is inherited Aave code under a lapsed BUSL; its own PSM and controller are AGPL-3.0.
Pros
- Close factor published precisely: 50% normally, rising to 100% at a health factor of 0.95 or below
- Spark PSM and ALM Controller released under AGPL-3.0, a strong copyleft open-source licence
- Audits named and dated: ChainSecurity, Cantina and Certora
- Parameters updated through the Sky Governance process
Cons
- SparkLend core carries a Business Source License naming Aave as licensor, inherited from the fork
- That licence's change date of 27 January 2023 has long passed, leaving its current status unclear on the face of the file
- No reserve factor or protocol fee figures are published
- No timelock duration is stated for Sky Governance
Radiant publishes the most specific liquidation terms in this comparison — a flat 15% penalty split evenly between liquidator and growth fund, with a 50% close factor — and places all contracts behind a 72-hour timelock. Its repository contains no licence file whatsoever.
Pros
- Liquidation penalty published as a flat 15%, split 7.5% to the liquidator and 7.5% to the Radiant Growth Fund
- Close factor stated: a liquidator may repay up to 50% of a single borrowed amount
- All contracts behind a timelock with a stated 72-hour delay, and the timelock addresses published
- Four auditing firms named: OpenZeppelin, BlockSec, Zokyo and PeckShield
Cons
- The official repository has no LICENSE file on either main or master — published code with no grant of rights
- Documentation states core contracts are upgradable by the owner
- Reserve factor and protocol fee percentages are not published
- Per-asset loan-to-value values are not documented
Compound III is controlled by the same Timelock contract that governs Compound v2, and its liquidation mechanics are documented in structure. Its BUSL change date of 31 December 2025 has passed while the licence file still reads BUSL, and no Comet audits could be found.
Pros
- All instances of Compound III controlled by the Timelock contract
- Liquidation structure documented: separate borrow and liquidation collateral factors
- Compound v2 remains BSD-3-Clause with an extensive published audit history
- v2 audits named: Trail of Bits, OpenZeppelin, Certora and Gauntlet
Cons
- The stated BUSL change date of 31 December 2025 has passed, yet the file still reads Business Source License
- No audit reports or audits directory could be found for Compound III (Comet)
- Timelock duration, voting period and proposal threshold are not stated in the v3 governance documentation
- Reserve factor and per-market collateral factors are not published
Fluid publishes a solid audit record from four named firms, but almost nothing else a borrower needs: fee values, collateral factors, liquidation penalties and governance are all left as contract fields with no documented values.
Pros
- Four auditing firms named across multiple reports: PeckShield, StateMind, MixBytes and Cantina
- Vault configuration structure is public, so the parameters that exist can be identified
- Contracts published openly for inspection
Cons
- Business Source License 1.1 whose grant permits educational use only and expressly prohibits production use
- Borrow fee, collateral factor, liquidation threshold and liquidation penalty exist as fields with no published values
- Governance and upgrade authority are not stated anywhere we could read
- No timelock is documented
At a glance
| Service | Score | Best for |
|---|---|---|
| Morpho | 8.0 | GPL-licensed with published liquidation curve |
| Venus Protocol | 7.4 | The most precisely documented governance delays |
| Kamino | 7.2 | Publishes what it charges, per market |
| Euler | 6.8 | Governance limits enforced permanently in code |
| Aave | 6.6 | 49 audit reports and a stated governance delay |
| Silo Finance | 6.2 | Liquidation fee fixed at deployment, forever |
| Spark | 6.2 | Publishes the close factor that most protocols hide |
| Radiant Capital | 6.0 | 72-hour timelock, no licence at all |
| Compound | 5.2 | Timelock-controlled, with a lapsed licence date |
| Fluid | 4.2 | Four audit firms, almost nothing else published |
A lending market lets you supply assets to earn interest or post collateral to borrow against. The advertised deposit rate is the least interesting number on the page: it moves with utilisation and can change block to block.
The terms that decide whether you keep your collateral are the loan-to-value limit, the liquidation threshold and the penalty applied when it is breached. We score those, the oracle that prices your collateral, and how the protocol has handled bad debt in the past.
How we score this category
In a lending market the terms that decide whether you keep your collateral are the liquidation threshold, the penalty and who can change them. This rubric therefore scores documentation rather than headline rates: rates move block to block and mean little, while a published liquidation penalty and a timelock on parameter changes tell you what happens on your worst day. Licences are read from the LICENSE file, not from the marketing page.
Every service in this ranking is scored 0–10 against the same 5 criteria, each carrying equal weight (20% apiece). The overall score is their plain average, rounded to one decimal place, and each review prints the individual numbers and the arithmetic so you can reproduce the result yourself.
What each criterion covers
- Fee disclosure20% of the score
- Whether protocol fees, reserve factors and any origination or exit charges are published as figures rather than described as configurable parameters. Stating there are no origination fees counts as disclosure; leaving the field undocumented does not.
- Licence & code openness20% of the score
- The licence in the repository, read directly. A recognised open-source licence scores highest, a delayed-open licence such as BUSL scores mid-table, and a repository with no licence file at all scores lowest — published code with no grant of rights reserves every right by default.
- Liquidation terms20% of the score
- Whether the liquidation threshold, close factor and penalty are published as numbers a borrower can plan around, rather than left as per-market parameters a reader must dig out of a contract.
- Governance & timelock20% of the score
- Who can change risk parameters, and whether a delay stands between a decision and its effect. A named timelock with a stated duration is the difference between a governed protocol and one that can be altered under you without warning.
- Audit disclosure20% of the score
- Whether audit reports are published with the firms named and dated. A long list of named firms with recent dates scores far above a claim that the protocol has been audited.
What does not affect a score
Nothing commercial, because there is nothing commercial to affect it. This is a non-commercial project: no advertising, no sponsored placements, and no affiliate or referral links anywhere on the site. Links to the services above earn us nothing, so no provider can buy a score, a rank, or an entry in this list. Where two services finish level, they are listed alphabetically rather than ordered silently.
The process across all categories, including how often we re-check, is set out in our ratings methodology.
Rankings are editorial. Nothing here is financial advice. Editorial policy · How we score · How we're funded.