Restaking protocols evaluated on exactly what becomes slashable and who decides, the worst-case exit rather than the advertised one, the incident record with dates and amounts, and whether the documentation still describes the live product.
EI
#1 of 7 · published ranking
EigenLayer (EigenCloud) native restaking
80ChainChoice Score
4199
Why it leads
Best in the pool on slashing surface (8/10; next 7/10)
Cost
Not priced· No comparable price is published
No provider can pay for a position in this table. The code that computes this order cannot read which links earn us a commission, and every build proves it. Every score below rebuilds from the published criteria.Ranking 2026.08 · 4 criteria · 7 products · same inputs, same order
Personal comparison
Restaking ranked comparison
No comparable price is published; ranking still uses verified product evidence.
RankProviderBest fitScore
1
EIEigenLayer (EigenCloud) native restakingThe base economic-security layer most o...Top ranked
Slashing surface — 8/10Trade-off: The 17.5-day figure is not your escape hatch, and this is the one thing to read befor...
Commission: This provider pays ChainChoice nothing for this link. Disclosure
Crypto products are high risk. Check the provider is authorised where you live before you deposit.
3
ether.fiether.fiThe category's exit, not its entry: eth...
Users who want liquid staking with a stronger onchain-utility posture after setup.Trade-off: Users who mainly want the simplest exchange-based staking route or a more native-post...
Commission: This provider pays ChainChoice nothing for this link. Disclosure
Crypto products are high risk. Check the provider is authorised where you live before you deposit.
1 more assessed and ranked below — none of them hidden for commercial reasons.
Ranking blindCommercial terms excludedDated primary sourcesCrypto products are high risk. Check the provider is authorised where you live before you deposit.
Ranked on 4 published criteria weighted 34/26/26/14, which set goal alignment — 30 of the 86 points. The rest: regional access 20, evidence depth 18, ease of use 10, institutional trust 8. Profile match is shown in breakdowns but carries no weight.
Audit
Infrastructure
Methodology
2026.09.15
published 2026-09-16
Providers tracked
980+
across 119 categories
Last verified
2026-08-17
newest dated pricing or sentiment read
Named on the roster
2 people
managing directors · 6 automated processes
Decision guide
What matters most before choosing in this category
Weighted on slashing surface (32), withdrawal path (26), security track record (24), and reward transparency (18). The heaviest is slashing surface: How many services the capital secures, and how clearly each slashing condition is disclosed.
What matters most before choosing in this category
Weighted on slashing surface (32), withdrawal path (26), security track record (24), and reward transparency (18). The heaviest is slashing surface: How many services the capital secures, and how clearly each slashing condition is disclosed.
Which networks is your capital allocated to — and can the wrapper add one without telling you?
What is the worst exit anyone has actually experienced, on the chain you are on?
Does the protocol document its own incidents anywhere a depositor would look?
Leads on disclosure
Expert review and scoring weights
EI
EigenLayer (EigenCloud) native restakingData checked Aug 2026
The base economic-security layer most of this category routes through, and the only member that publishes its slashing and withdrawal parameters as numbers you can read straight off the deployed contracts — 126000 / 100800 / 100800 / 50400 blocks, all four re-confirmed onchain on 2026-08-16. Native ETH is also the one strategy EigenLayer refuses to let an AVS redistribute, so a native restaker's downside is destruction, never transfer to a named third party. The cost is that you must pick an operator, and the operator — not the protocol — sets how much notice you get before your ETH becomes slashable somewhere new.
The base economic-security layer most of this category routes through, and the only member that publishes its slashing and withdrawal parameters as numbers you can read straight off the deployed contracts — 126000 / 100800 / 100800 / 50400 blocks, all four re-confirmed onchain on 2026-08-16. Native ETH is also the one strategy EigenLayer refuses to let an AVS redistribute, so a native restaker's downside is destruction, never transfer to a named third party. The cost is that you must pick an operator, and the operator — not the protocol — sets how much notice you get before your ETH becomes slashable somewhere new. Strongest on what exactly can be taken from me, and who gets to decide? (8/10): Four of five elements are published as machine-readable parameters; I re-read five of them off the deployed mainnet contracts myself rather than trusting the docs. (b) MAX FRACTION — exact. AllocationManager.md: "`wadsToSlash` MUST be within the bounds `(0, 1e18]`" (line 954), i.e. Weakest on does the wrapper tell me what it runs on, and is what it tells me still true today? (5/10): The numbers are exactly right; the pages around them are not. Score unchanged at 5, but for materially different reasons than the draft gave: I found the fee disclosure the draft said was missing, and three currency defects the draft missed. WHAT SURVIVED A SAME-DAY CHECK — the strongest single finding in this record. Published price: EigenLayer publishes no deposit, withdrawal or management fee for native restaking on any restaker-facing page.
Best forSlashing surface — 8/10
Main tradeoffThe 17.5-day figure is not your escape hatch, and this is the one thing to read before committing. ALLOCATION_CONFIGURATION_DELAY (126000 blocks) is the delay before an operator can CHANGE their notice period — it is not the notice you get. The window that actually protects you is ALLOCATION_DELAY, which EigenLayer's own reference table says is set by the operator and on which "The protocol provides no constraints", and which its contract documentation says "operators can choose to configure this delay however they want - including setting it to 0." Two further paths give you no notice at all: delegating to an operator who is already allocated makes "all funds of the `staker` … immediately slashable" on the spot, and an AVS adding a strategy to an existing operator set makes "any existing allocations to that strategy instantly slashable." So the security of your ETH here is a property of the operator you choose, not of the protocol you deposit into — and no vendor page enumerates the live operator sets: the app that would is returning HTTP 500 today, and the ecosystem page that might substitute now lists AI-agent demos with not one occurrence of the word AVS. The compensating fact, and the reason native restaking is the more conservative way to use this platform: native ETH is the one strategy an AVS cannot point at a redistribution address. "Native ETH and EIGEN are not yet eligible for redistribution" — so your worst case is that the ETH is destroyed, never that it is paid to whoever chose to slash you. Budget ~31 days for a full exit, not the 8-17 days one docs page still advertises off a superseded 7-day escrow; the escrow is 14 days, published on four other vendor pages and confirmed onchain.
Verify before signupEigenLayer publishes no deposit, withdrawal or management fee for native restaking on any restaker-facing page. It DOES publish the fee taken from rewards, in plain language, on its rewards concept pages: "Operators can [set their per-AVS fee rate](../../operators/howto/configurerewards/set-rewards-split.md) on AVS Rewards to any amount from 0% to 100%. The default split is 10%." (concepts/rewards/rewards-concept.md), and "By default, Operators earn a 10% split on Rewards. The rest of the reward is claimable by the Operator's delegated Stakers." (concepts/rewards/rewards-split.md). The same 10% default is stated for programmatic incentives (concepts/rewards/pi-split.md). Note one internal contradiction: concepts/rewards/rewards-claiming-faq.md still says "Operators get a fixed 10% portion rewards, though this is subject to change in a future release to be variable", which the 0–100% concept page has already superseded. The contract docs add a mechanical restatement — "The split lookup chain is: operator set split → global default split → 10% fallback." (docs/core/RewardsCoordinator.md) — and mention an unquantified EigenLayer-level charge, "Takes the protocol fee (if the submitter is opted in for protocol fees)", without ever stating its rate. Checked 2026-08-16: nothing under docs/eigenlayer/restakers/ states a rate; the fee disclosure lives one directory over in concepts/rewards/. Native restaking is otherwise gas-cost bearing: "the web app will batch up to 80 validators per proof transaction batch."
Methodology
How this category is reviewed
Reviewed on what can be taken and who decides, the worst-case exit rather than the advertised one, the incident record with dates and amounts, and whether the vendor’s own documentation still describes what ships. Weights are the engine’s: 34 / 26 / 26 / 14.
The order on this page is the published ranking for this category. Every criterion, weight and source behind it is on the methodology page.
Frequently asked
Questions people ask before choosing restaking
What exactly can be taken from me?
That depends on whether you hold the base layer or a wrapper. The base platforms publish slashing parameters machine-readably, so the venues, the maximum fraction and the slashing key are all checkable before you deposit. Several liquid wrappers do not, and the sharper question for those is whether they can add a new slashing venue without depositor consent or notice. A wrapper that chooses your venues and does not name them is asking you to accept a risk it has not described.
Is ether.fi still a restaking product?
Largely not, and this is the most important thing in the category. In August 2026 ether.fi hardened weETH with what it describes as "a cryptographic hard-disable on slashing exposure, with no onchain re-enable" — so its flagship token is now plain liquid staking. Restaking moved to a separate, much smaller token. Its own documentation has not caught up and still calls weETH a "liquid restaking token", and the homepage still describes the protocol as "a decentralized, non-custodial liquid restaking protocol". If you want ETH exposure with the least added risk, that is the answer; just do not buy weETH believing it is restaking.
What happened in the $292M exploit, and is it resolved?
It is resolved, and the interesting part is what came after. In April 2026 roughly 116,500 rsETH was drained through a bridge configured with a single verifier — no second party had to agree. Recovery was announced within a month, withdrawals and bridging resumed, funded by an industry restitution raise, and the protocol now carries several hundred million in TVL and processes exits normally. What it does not carry is any incident notice on a page it controls: the live marketing line is "battle-tested, institutional grade", and the documentation that described the affected system now returns 404. Surviving the year’s largest hack is not the mark against it. Not saying so is.
Why does the ranking disagree with how well-known these are?
Because the axes reward disclosure rather than size. The base layers rank highest on what can be slashed because they publish it in a form you can verify; the wrappers that state an explicit worst-case exit rank highest there; and the largest incident in the category costs its protocol on the record axis not for being hit but for documenting nothing afterwards. A brand you recognise is evidence that other people deposited, which is not the same as evidence that you can find out what you are accepting.
Not financial advice · For informational purposes only · Always do your own research
//Analytics consent·GDPR · ePrivacy · TTDSG
ChainChoice measures page views and conversions with two cookieless, EU-hosted services: Plausible and Cloudflare Web Analytics. Nothing loads until you accept, and rankings are identical either way.ChainChoice measures how the engine is used — page views, conversions, referrer — through two cookieless, EU-hosted services: Plausible and Cloudflare Web Analytics. No advertising cookies, no cross-site profile, no data resale. Neither script loads until you accept, and rankings are identical whether you accept or decline.
