Ranked & Reviewed
Best Liquid Restaking Protocols
Restaking protocols ranked on the extra slashing risk taken, commission, peg behaviour and how plainly risks are disclosed.
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.
Renzo charges 10% on restaking rewards, split evenly between treasury and node operators, passes 100% of EigenLayer rewards through, and documents its withdrawal delays layer by layer. ezETH is value-accruing, the code is BUSL-1.1, and four audit firms are named.
Pros
- Fee published as 10% on restaking rewards, split 50/50 between treasury and node operators
- States it passes 100% of EigenLayer rewards through
- Withdrawal delays documented by layer: 9-10 day beacon exit, 14 day EigenLayer minimum, 3 day Renzo buffer
- ezETH is value-accruing, like a cToken
- Four auditors named with reports: Halborn, Sigma Prime, Code4rena and Nethermind
Cons
- The risk page lists slashing among integrated-protocol risks but does not detail the additional restaking slashing conditions
- No insurance is mentioned
- Business Source License 1.1 until 1 December 2026
YieldNest states outright that AVSs can impose additional slashing conditions on restaked ETH, releases ynETH under BSD 3-Clause, and names four auditors. Its protocol fee and withdrawal fee are described as dynamic and embedded, without a published percentage.
Pros
- States plainly that AVSs can impose additional slashing conditions on restaked ETH or ERC20 tokens
- BSD 3-Clause licence — a recognised open-source licence
- Four auditors named with reports: Zokyo, Chain Security, Composable Security and NFR Audits
- ynETH is value-accruing and auto-compounding, non-custodial, with an Immunefi bug bounty
Cons
- The protocol fee is described as a dynamic structure with no fixed percentage published
- A withdrawal fee is embedded in the claim NFT but the percentage is not specified
- Exit depends on the beacon-chain queue plus EigenLayer's 7-day delay
ether.fi splits staking rewards 90% to stakers, 5% to node operators and 5% to the protocol, offers both a rebasing eETH and a value-accruing weETH, and licenses its contracts under MIT. Its slashing and audit disclosures could not be verified.
Pros
- Reward split published: 90% stakers, 5% node operators, 5% protocol
- Two token forms: eETH rebasing and weETH wrapped and value-accruing
- Native restaking on EigenLayer at the protocol level
- MIT licence on the smart contracts, stated in the repository
Cons
- Slashing conditions and any insurance could not be verified from a fetched official page
- No audit page could be fetched and no firm is verified
- Withdrawal has protocol-level waiting periods, subject to available liquidity
Mellow builds restaking vaults on both Symbiotic and EigenLayer, with a fee manager defining deposit, redemption, performance and protocol fees as on-chain parameters. Six audit firms are named, but the fee figures, unbonding period and slashing detail are not published as fixed values.
Pros
- Integrates both Symbiotic and EigenLayer through dedicated verifier modules
- Six auditors named with reports: StateMind, ChainSecurity, Sherlock, MixBytes, Nethermind and Decurity
- Fee structure documented in four categories, all paid in vault shares
- Business Source License 1.1, converting to GPL in 2028
Cons
- Fee percentages are on-chain parameters, not published as fixed figures
- The unbonding period and any exit fee are not stated on the pages read
- Slashing and insurance detail was not verified on the pages read
- The receipt token's rebasing-versus-accruing model is not explicitly stated
Symbiotic is restaking infrastructure rather than a single token: networks define slashing conditions against collateral, and its risk documentation stresses that conditions should be transparent and proportional. It names twelve auditors and a $500,000 bug bounty, under a BUSL licence.
Pros
- Twelve auditors named including OpenZeppelin, Sigma Prime, Sherlock, Certora, ChainSecurity and Zellic
- A $500,000 bug bounty via Cantina
- Risk framework published, stressing that slashing conditions should be clearly defined, transparent and proportional
- Business Source License 1.1, converting to GPL in July 2026
Cons
- It is infrastructure, not a consumer LRT — there is no single fee or token to score for a holder
- No protocol fee is stated
- The unbonding period and collateral-token mechanics were not on the pages that loaded
- Slashing depends on each network's own conditions rather than a single protocol rule
Kelp charges a 10% fee on rewards earned on ETH deposits for rsETH, with a 7 to 10 day withdrawal window, and licenses its code under BUSL-1.1 converting to GPL v3 in 2028. Its token mechanics, slashing detail and audits could only be partly confirmed.
Pros
- Fee published as 10% on the rewards earned on ETH deposits
- Withdrawal window stated as 7 to 10 days
- Business Source License 1.1, converting to GPL v3 on 31 December 2028
- Sigma Prime and Code4rena named as auditors in Kelp's documentation
Cons
- The receipt token's rebasing-versus-accruing model could not be verified
- Slashing conditions and any insurance were not verified from a fetched page
- The GitBook documentation and audits page returned 404 on direct access, so several facts came via search extraction
Karak secures Distributed Secure Services that can slash restaked funds, documents a 9-day minimum withdrawal delay with a 7-day slashing window, and is BUSL-1.1 converting to MIT in 2027. Its documentation site fails TLS, so most facts came via search extraction and an audit mirror.
Pros
- Distributed Secure Services model documented, with the DSS able to slash malicious behaviour
- Withdrawal delay stated: a 9-day minimum, combining a withdraw window and a veto window
- Slashing window stated at up to 7 days before withdrawal initiation
- Per-asset ERC-4626 vaults, a value-accruing share model
- BUSL-1.1 converting to MIT on 2 April 2027, read from an audit mirror of the contracts
Cons
- The documentation site fails TLS (certificate covers only Netlify), so most facts came via search extraction
- No fee figure was found
- No insurance or exit fee is stated
- Only a Code4rena audit is identifiable, via a third-party platform rather than Karak's own page
Swell's rswETH charges a 10% fee after a zero-fee launch period, is reward-bearing, and points validators directly at EigenLayer to secure AVSs. Its withdrawal delays are documented, but the core repository has no licence file and slashing is not addressed.
Pros
- Fee published as 10% following a zero-fee first 30 days
- rswETH is reward-bearing, appreciating as staking and restaking rewards accrue
- Points validators directly at EigenLayer smart contracts to secure AVSs
- Withdrawal delays documented: 9-16 day exit queue plus EigenLayer's 7-day delay, buffer enables ~1 day
- Sigma Prime named as auditor, with Gauntlet and Chaos Labs as risk firms
Cons
- The core repository has no top-level LICENSE file, and per-file SPDX headers could not be verified
- Slashing conditions are not addressed on the fetched page
- No insurance is mentioned
EigenLayer is the restaking protocol most others here build on. It documents opt-in slashing, live since April 2025, a 14-day withdrawal delay, and that natively restaked ETH is permanently locked in EigenPods when slashed. Its docs were largely blocked, so several facts came via search extraction.
Pros
- Slashing documented as opt-in, live on mainnet since the April 2025 ELIP-002 upgrade
- States plainly that natively restaked ETH cannot be redistributed and is permanently locked in EigenPod contracts when slashed
- Withdrawal delay stated at 14 days after queuing
- Business Source License 1.1, converting to GPL v2-or-later on 19 July 2026
Cons
- No fee figure was found on any reachable page
- The documentation site returned 403 on every direct fetch, so most facts came via search extraction
- No audit record could be verified from a fetched official page
- It issues no LRT itself; restakers deposit LSTs or native ETH for delegated shares
Eigenpie offers isolated liquid restaking — a separate value-accruing token such as egETH per underlying LST, to ring-fence each asset's risk — restaked through operators on EigenLayer. Beyond that structure, its fee, withdrawal terms, licence and audits could not be verified.
Pros
- Isolated restaking model documented: a separate mLRT per LST, isolating each asset's risk
- egETH is value-accruing, using an underlying share system that appreciates against ETH
- Restakes through node operators on EigenLayer to support AVSs
Cons
- No fee percentage is stated on the fetched pages
- Slashing conditions and any insurance are not stated
- Withdrawal terms — unbonding period and any exit fee — are not detailed
- No official public contracts repository could be located, so the licence is unverifiable
- No audits are named on the fetched pages
At a glance
| Service | Score | Best for |
|---|---|---|
| Renzo (ezETH) | 7.6 | Fee split and withdrawal delays published in full |
| YieldNest (ynETH) | 7.0 | BSD-licensed with slashing stated plainly |
| ether.fi (eETH / weETH) | 5.6 | MIT-licensed, three-way reward split published |
| Mellow | 5.6 | Six named auditors on a Symbiotic-and-EigenLayer vault |
| Symbiotic | 5.4 | Twelve named auditors and a stated risk framework |
| Kelp DAO (rsETH) | 5.0 | 10% fee and a delayed-open licence |
| Karak | 4.8 | Nine-day withdrawal with a slashing window, docs blocked |
| Swell Restaking (rswETH) | 4.8 | Value-accruing rswETH, no licence file |
| EigenLayer | 4.6 | The restaking layer itself, opt-in slashing stated |
| Eigenpie (mLRT) | 2.0 | Isolated per-LST restaking, little else documented |
Restaking reuses staked capital to secure additional services, earning a second layer of rewards. It also accepts a second layer of slashing conditions, which is the part most marketing leaves out.
We score what is actually being secured and under what penalties, how losses would be allocated to depositors, and whether the protocol separates genuine yield from points or incentive programmes that may never convert into anything.
How we score this category
Restaking layers extra slashing conditions on top of staking, so the yield you take on carries a risk plain staking does not. This rubric scores whether a protocol names its commission, explains that additional slashing surface rather than glossing over it, states its exit terms, and licenses its code openly. A protocol that documents the new risk honestly scores above one that markets only the extra yield.
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
- Commission clarity20% of the score
- Whether the fee on restaking rewards is published as a figure and split out. A stated percentage, ideally with where it goes, beats a dynamic fee described only as a parameter.
- Slashing & risk disclosure20% of the score
- Whether the additional slashing conditions restaking introduces are explained, what is being secured, and whether any insurance exists. Naming the extra risk plainly is the single most important thing a restaking protocol can document.
- Exit & redemption20% of the score
- The unbonding period — which stacks the beacon-chain exit queue on top of the restaking-layer delay — whether an instant route exists, and any exit fee.
- Licence & code openness20% of the score
- The licence on the protocol's own repository, read from the file. A recognised open-source licence scores highest; a delayed-open licence mid-table; a missing licence file lowest.
- Audit disclosure20% of the score
- Whether audit reports are published with the firms named. A long list of named firms scores far above a claim of having been audited by leading firms with none named.
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.