Compound Review
One documented timelock over every Compound III market
of 10
Compound III scores 5.6 among lending protocols, with a 7 for governance and no mark higher. No fee or reserve figures are published and its two named audits are undated in the docs. Re-scored from 5.2 on desk research checked November 1, 2026.
Overall score
5.6 = (3 + 6 + 6 + 7 + 6) / 5; was 5.2
Licence
BUSL-1.1; change date Dec 31, 2025 passed; change licence GPL-2.0-or-later
Timelock
One Timelock for all markets; 2 days after queuing; local delay off Ethereum
Governance votes
25,000 COMP to propose; 3-day vote; 400,000 votes in favor to pass
Audits
OpenZeppelin (July 2022) and ChainSecurity, linked from the docs
Liquidation
Absorb by any address; penalty set per asset by its liquidation factor
Scores
Each criterion carries equal weight. The overall score is the average of these 5 scores — (3.0 + 6.0 + 6.0 + 7.0 + 6.0) ÷ 5 = 5.6.
Pros
- Governance 7: one Timelock controls all Compound III markets, with a 2-day delay and 400,000-vote quorum
- Liquidation process documented by formula, from absorb to the buy-collateral discount
- Institutional Market lists LTV of 81%-87% and penalties of 5%-10% by asset
- Licence 6: the BUSL change date of December 31, 2025 has passed, granting GPL-2.0-or-later rights
Cons
- Fee disclosure 3: no reserve, kink or rate figures are published for any market
- Per-asset liquidation factors for the main markets are not in the documentation
- Audits 6: only two firms named, undated in the docs, and no audits folder in the repository
- Licence file text still reads BUSL, so its converted status is not stated on its face
- Governance figures sit in the v2 docs, and a multisig Pause Guardian can halt operations outside the vote
How Compound scores in crypto lending
This entry rates Compound III, also called Comet, the current version of the Compound lending protocol, in our lending ranking. The basis is desk research of the Compound documentation (now at docs.compound.xyz), the Comet repository and the Foundation's market pages, checked on November 1, 2026. No position was opened. The category uses five criteria, each marked 0 to 10 and weighted equally, as described in our ratings methodology.
| Criterion | What the sources show | Mark |
|---|---|---|
| Fee disclosure | Rates come from governance-set curves and reserves from the borrow-supply spread. No reserve or fee figures are published; the interface charges no fee. | 3 |
| Licence & code openness | BUSL-1.1 with a change date of December 31, 2025, after which the file's own terms grant rights under GPL-2.0-or-later. The file text is unchanged. | 6 |
| Liquidation terms | The absorb and buy-collateral process is documented by formula. Per-asset figures appear only on the Institutional Market page. | 6 |
| Governance & timelock | One Timelock controls every Compound III market, with an extra local delay on other networks. The 2-day delay, 25,000 COMP threshold and 400,000-vote quorum are stated in the v2 docs. | 7 |
| Audit disclosure | The docs name OpenZeppelin and ChainSecurity audits of Compound III, without dates; OpenZeppelin's report is from July 2022. The repository has no audits folder. | 6 |
The overall score is the plain average: (3 + 6 + 6 + 7 + 6) ÷ 5 = 28 ÷ 5 = 5.6. That keeps Compound ahead of Fluid and below Silo Finance at 6.0.
Two marks were re-scored from current sources on this check. Licence and code openness rose from 5 to 6 after the licence's December 31, 2025 change date passed, the same mark Spark carries for a lapsed BUSL date. Audit disclosure rose from 5 to 6 because the documentation now links two named Compound III audits, where the previous check found none. The overall score moves from 5.2 to 5.6.
Strongest criteria: governance and liquidation mechanics
Governance & timelock (7)
The Compound III governance page states that all instances are controlled by the same Timelock contract that administers Compound v2. Markets on networks other than Ethereum add a local timelock with a further delay before admin functions run. A Pause Guardian, designated by COMP holders and held by a community multisig, can pause supply, transfers, withdrawals, absorptions and collateral purchases. The numbers sit one level away, in the v2 governance docs: a 25,000 COMP proposal threshold, a three-day vote, at least 400,000 votes in favor, and implementation two days after queuing. The mark reflects a real delay on every change, with the detail split across two sets of pages and a pause power held outside the voting process.
Liquidation terms (6)
The liquidation page explains the mechanism clearly. An account becomes liquidatable once its debt exceeds limits set by liquidation collateral factors, which sit above the borrow collateral factors to leave a buffer. Anyone can call absorb. The protocol takes over the debt and collateral and credits the user with the collateral's value minus a penalty set by each asset's liquidation factor. If reserves are below target, liquidators can buy the seized collateral at a discount set by a stated formula. The general docs give no per-asset values. The Foundation's Institutional Market page does, for one USDC market: loan-to-value of 81% to 87%, liquidation factors of 86% to 93% and penalties of 5% (ETH) to 10% (wBTC and cbBTC). For the main markets, a borrower has to read the figures from the contracts.
Weakest criterion and the middle marks
Fee disclosure (3): a deficiency
Compound does not charge a separate protocol fee. Its cost to users is the gap between borrow and supply rates, which builds reserves. Both rates come from governance-set curves with a utilization kink. The documentation gives the model but no reserve rates, kink points or current figures for any market. The Institutional Market page lists collateral terms but no fee or reserve figures either. The rubric asks for figures rather than configurable parameters, so this is a measured gap. Our guide to reading a DeFi lending market shows where to find live rates instead.
Licence & code openness (6): converted on its own terms
The Comet LICENSE file is Business Source License 1.1, with Compound Labs as licensor. Its change date is the earlier of December 31, 2025 or a date published at a named ENS record, and the change licence is GPL-2.0-or-later. The terms in the same file say that from the change date the licensor grants rights under the change licence. The file was last edited in February 2025, so it still reads BUSL. Read on its own terms, the code has carried GPL rights since January 2026, but the file a reader opens still presents a source-available licence. The category scores a lapsed BUSL date with an unedited file at 6, between delayed-open licences (3 to 4) and plainly stated open source licences (9 to 10). Updating the file to the GPL text would remove the ambiguity.
Audit disclosure (6)
The earlier finding that no Comet audits could be found no longer holds. The documentation's security section links two Compound III audits, by OpenZeppelin and ChainSecurity, and mentions a bug bounty. OpenZeppelin's report was published in July 2022 and found no critical issues. The docs date neither report, the Comet repository has no audits folder, and no later review of the many markets added since appears on the page. Two named firms is enough to lift the mark off the floor the earlier finding implied. An undated, four-year-old record with no newer reviews keeps it below peers that publish three or more dated reports.
What the score does not cover
- Rates and rewards. Supply and borrow rates and COMP reward rates were not measured.
- Compound v2. The older protocol, under BSD-3-Clause, still has its own app and is not rated here.
- Collateral yield. Supplied collateral earns no interest in Compound III; this design choice is not scored.
- Networks. Which markets the app shows on each network is outside the criteria.
Compound vs Silo Finance, Spark and Fluid
Spark scores 6.2, six tenths above Compound. It shares Compound's licence mark (6), and leads on fees (4 against 3), liquidation terms (8 against 6) and audits (7 against 6), while trailing on governance (6 against 7). Silo Finance, at 6.0, leads through fees (6) and a 9 for audits, which outweigh its 3 for licence and 5 for governance. Fluid, at 5.0, matches Compound on fees and beats it on audits, 9 against 6. It falls behind on licence (3), liquidation terms (5) and governance (5). Compound's middle-of-the-scale marks across the board, with no criterion above 7, explain its place near the foot of the table.
Who this rating suits
The score suits readers who want a single, documented timelock over every market and a liquidation process explained by formula. Readers who need fee and reserve figures in the documentation, or a dated, recent audit record, should weigh the lower marks. Borrowing against volatile collateral can end in liquidation; the rating describes published documentation and is not advice to lend or borrow.
How it compares
Morpho
Open-source code and liquidation terms fixed by formula
Venus Protocol
Dated audit catalog and timelocks stated to the hour
Kamino
Publishes its fee spread per market and states no other fees
Euler
Dated, linked audit record and fee limits enforced in code
Aave
Deep, dated audit archive and documented governance delays
Spark
Publishes the health factor at which full liquidation applies
Silo Finance
Liquidation fee fixed for each market's lifetime
Fluid
Audited core: seven reports from four named firms
Sources
Scored from published primary sources rather than hands-on use. Every figure above comes from one of the pages below, on the date shown. Manufacturers change prices and specifications without notice — if something here no longer matches the source, tell us and we will correct it.
- 1.Compound III (Comet) — LICENSE — checked
- 2.Compound docs — overview and security — checked
- 3.Compound docs — governance (Compound III) — checked
- 4.Compound docs — governance (v2) — checked
- 5.Compound docs — liquidation — checked
- 6.Compound docs — interest rates — checked
- 7.Compound Foundation — Institutional Market — checked
- 8.OpenZeppelin — Compound III audit (July 2022) — checked
Frequently asked questions
Why does Compound score 5.6?
It is the plain average of five equal marks: 3 for fee disclosure, 6 for licence and code openness, 6 for liquidation terms, 7 for governance and timelock and 6 for audit disclosure. They total 28, and 28 divided by 5 is 5.6.
What holds Compound's score back?
Mainly fee disclosure. The docs describe the rate model and reserves but publish no fee or reserve figures. The audit record is also thin: two firms, undated in the docs, with nothing newer listed.
Is Compound III open source now?
The licence file still reads BUSL-1.1, but its change date of December 31, 2025 has passed, and its own terms grant GPL-2.0-or-later rights from that date. Because the file text was not updated, the licence marks 6 rather than the 9 to 10 given to plainly stated open source licences.
What changed since the previous score of 5.2?
Two marks rose by one point each. Licence and code openness went from 5 to 6 after the BUSL change date passed, and audit disclosure from 5 to 6 because the docs now link two named Compound III audits.
Which criteria were actually measured?
Published fee figures, the licence file, documented liquidation terms, timelock and governance rules, and named audits, all read on November 1, 2026. Rates, rewards and Compound v2 were not scored.
How does Compound compare with Spark and Fluid?
Spark scores 6.2, level on licence and ahead on fees, liquidation terms and audits. Fluid scores 5.0: it beats Compound on audits, 9 against 6, but trails on licence, liquidation terms and governance.
Nothing here is financial advice. Editorial policy · How we score · How we're funded.