EigenLayer Review: Restaking Protocol Analysis
Introduction
EigenLayer changed how Ethereum's security model works in practice. Before restaking, every new middleware service — oracles, bridges, data availability layers, rollup sequencers — had to bootstrap its own validator set from scratch, producing a patchwork of isolated security pools far weaker than Ethereum's consensus layer. EigenLayer lets you reuse your already-staked ETH as security collateral for those services, creating a shared security marketplace.
Since the AVS marketplace opened in April 2024, EigenLayer has attracted staked ETH from both solo validators and liquid staking token holders into an Autonomous Verifiable Services (AVS) ecosystem spanning data availability layers, oracle networks, bridge security modules and rollup sequencing. It has also passed its high-water mark: DeFiLlama records restaked value peaking near $22 billion in August 2025 and falling to roughly $5 billion by August 2026. That contraction is the single most important piece of context for anything else you read about restaking, including this review, and we treat it as a fact to explain rather than a detail to bury.
In this review we analyse EigenLayer's restaking architecture, its AVS ecosystem and reward mechanics, its security posture including slashing, and how to choose an operator — with every figure dated, because most of them have moved a long way since 2024.
Our rating of 4.3/5 reflects EigenLayer's strong technical foundation and clean protocol-level record since 2024, balanced against the complexity, the shrinking demand side and risk layers that restaking introduces for you. The protocol is a genuine innovation in blockchain security design, but it demands more sophisticated risk management from you than simple staking, for a premium that has narrowed considerably. For a broader view of how EigenLayer fits into the liquid staking yield landscape, see our liquid staking yield strategies guide.
What Is EigenLayer
EigenLayer is a middleware protocol built on Ethereum that introduces the concept of restaking — the ability for you to reuse your staked ETH as cryptoeconomic security for services beyond Ethereum's consensus layer. Founded by Sreeram Kannan, a professor at the University of Washington, the protocol addresses a fundamental inefficiency in blockchain infrastructure: every new decentralised service traditionally needed to recruit and incentivise its own set of validators, duplicating security costs across the ecosystem.
The core insight behind EigenLayer is that if you are an Ethereum validator, you have already committed significant capital (32 ETH per validator) to secure the network. This committed capital represents a massive pool of cryptoeconomic security — close to 40 million ETH staked network-wide as of mid-2026 — that sits idle relative to its potential. EigenLayer allows your capital to simultaneously secure additional services — called Autonomous Verifiable Services in EigenLayer's current documentation, and Actively Validated Services in older material, but abbreviated AVS either way — without requiring you to unstake from Ethereum. The result is a shared security marketplace where AVS builders can access Ethereum-grade security at a fraction of the cost of bootstrapping their own validator networks.
Core Concept: Shared Security
The shared security model works through a delegation mechanism. As a staker (either a solo validator or LST holder), you opt into EigenLayer by restaking your assets, then delegate to operators who run the validation software for various AVS. Each AVS defines its own validation tasks, reward structures, and slashing conditions. Operators who successfully validate earn rewards from the AVS, which are distributed to you after commission.
This creates a three-sided marketplace: you provide capital, operators provide infrastructure, and AVS builders consume security. The security available to any individual AVS scales with the restaked ETH allocated to it, which is why the direction of restaked value should matter to you as much as its level.
Critically, EigenLayer does not require you to move your ETH off the Ethereum beacon chain. If you are a native restaker, you maintain your validator duties whilst simultaneously opting into AVS validation through EigenLayer's smart contracts. If you are an LST restaker, you deposit your liquid staking tokens (stETH, rETH, cbETH) into EigenLayer's strategy contracts, which then make your assets available as security collateral for AVS. In both cases, your underlying ETH continues earning base staking rewards from Ethereum consensus.
Protocol History and Development
If you are evaluating EigenLayer's maturity, you should understand its development timeline. The project began in 2022, with the whitepaper published by Sreeram Kannan outlining the theoretical framework for restaking. The project raised $50 million in a Series A round led by Blockchain Capital in early 2023, followed by a further $100 million round in early 2024. This substantial funding should give you confidence that the team has the resources to maintain and develop the protocol long-term.
The protocol launched in stages: LST restaking deposits opened in June 2023 under caps that were gradually raised, native restaking via EigenPods followed in late 2023, and the AVS marketplace with operator delegation went live in April 2024 alongside EigenDA, the first production AVS. By the end of 2024, over a dozen AVS were live on mainnet, and restaked value went on to peak at around $22 billion in August 2025 before falling back sharply.
Throughout 2025, EigenLayer focused on expanding the AVS ecosystem, improving operator tooling, and shipping the slashing mechanism, which had been delayed from the initial launch and finally went live in 2025. You should note that the EIGEN token, transferable since September 2024, handles intersubjective faults — misbehaviour that cannot be proven on-chain but that reasonable observers would agree deserves a penalty — through a forking mechanism rather than ordinary slashing. As of August 2026, on-chain records show 20 distinct AVS addresses have created live Operator Sets under the current slashing model, against 56 that ever registered under the older AVSDirectory. Marketing counts of "190+" or "200+" AVS include testnet and in-development services.
Restaking Architecture Deep Dive
EigenLayer's architecture consists of several interconnected smart contract systems that manage your restaking deposits, operator delegation, AVS registration, and reward distribution. Understanding these components is essential for you to evaluate the protocol's security properties and make informed restaking decisions.
Native Restaking vs LST Restaking
EigenLayer supports two distinct restaking paths, each with different technical requirements, risk profiles, and yield characteristics:
Native Restaking is designed for you if you run your own Ethereum validator nodes. You create an EigenPod — a smart contract that serves as the withdrawal address for your Ethereum validator. When your validator's withdrawal credentials point to your EigenPod, EigenLayer can verify your validator's beacon chain balance and use it as restaking collateral. This approach provides you with the highest capital efficiency because your same ETH simultaneously secures Ethereum consensus and EigenLayer AVS without any intermediary tokens.
The technical process involves you deploying an EigenPod contract through EigenLayer's PodManager, then setting your validator's withdrawal credentials to the EigenPod address. EigenLayer verifies your validator's balance through beacon chain state proofs, which are submitted periodically to confirm your restaked amount. You should note that native restaking requires running validator infrastructure and managing withdrawal credentials, making it more suitable if you are technically sophisticated or an institutional staker.
LST Restaking allows you to deposit your liquid staking tokens — stETH (Lido), rETH (Rocket Pool), cbETH (Coinbase), and several others — into EigenLayer's StrategyManager contracts. Your deposited LSTs serve as restaking collateral whilst continuing to accrue their underlying staking rewards for you. This path is significantly simpler than native restaking, requiring only a standard token deposit transaction from you, and is accessible regardless of your technical expertise.
The trade-off you should consider is that LST restaking introduces an additional smart contract layer (the LST protocol itself) between you and the underlying staked ETH. This means you face both EigenLayer smart contract risk and LST protocol risk simultaneously. The early deposit caps that once throttled LST restaking are gone: EigenLayer now runs permissionless token strategies, so any ERC-20 — not just the headline LSTs — can be added as restaking collateral.
Delegation Model and Operator Roles
Once your ETH is restaked (either natively or via LSTs), you delegate your stake to operators. Operators are entities that register with EigenLayer and commit to running the validation software for one or more AVS. The delegation model is non-custodial — operators never take custody of your restaked assets. Instead, they receive the right to use your delegated stake as security collateral for AVS validation, and in return, they distribute rewards to you after deducting a commission.
Operators must register with EigenLayer by providing their Ethereum address, metadata URI (containing information about their infrastructure and policies), and initial AVS opt-in selections. Once registered, operators can accept your delegation and begin validating AVS tasks. Each operator independently decides which AVS to validate based on their infrastructure capabilities, risk appetite, and expected rewards.
The delegation relationship is one-to-one: you delegate your entire restaked balance to a single operator. This simplifies the accounting and slashing mechanics but means you must choose your operator carefully. If your operator is slashed for misbehaviour on an AVS they validate, the penalty falls on the stake they allocated to that Operator Set, and you take a pro-rata share of it. An AVS may take up to the whole of that allocation, but it cannot reach stake the operator has not allocated to it — so your exposure tracks your operator's allocation decisions, which they can change. That makes operator choice the decision to get right — see the operator selection guide below.
Operators set their own commission on the rewards they generate for you — anywhere from 0% to 100%, varying per AVS, with a default split of 10% to the operator. The rate is visible on-chain, so you can compare operators on both performance and cost before delegating.
AVS Ecosystem and Rewards
The Autonomous Verifiable Services ecosystem is the demand side of EigenLayer's marketplace — these are the protocols and services that consume your restaked security. The health and growth of the AVS ecosystem directly determines the yield available to you, making it a critical factor in your evaluation of EigenLayer's long-term value proposition.
Active AVS Services
As of August 2026, around 20 AVS have live Operator Sets on mainnet, spanning several categories of blockchain infrastructure that you can earn rewards from:
Data Availability: EigenDA is EigenLayer's flagship AVS. It provides a high-throughput data availability layer for Ethereum rollups, competing with Celestia and Ethereum's native danksharding. EigenDA leverages your restaked ETH to guarantee data availability for rollup transactions, with operators storing and serving data blobs. Several major rollups have integrated EigenDA for their data availability needs, generating consistent demand for your restaked security.
Oracle Networks: Oracle AVS provide price feeds, randomness and cross-chain data. The attraction is that the cost of corrupting a feed scales with the restaked ETH backing it rather than with the oracle's own token market cap, which makes it dearer to attack than a standalone network running on a small security budget.
Bridge Security: Cross-chain bridge AVS secure message passing and asset transfers between chains. Bridges have historically been the most exploited category of DeFi infrastructure, so backing them with restaked ETH rather than a small multisig committee is a genuine improvement — Kelp's April 2026 bridge exploit is a reminder of what the alternative costs.
Rollup Sequencing and Verification: Several AVS provide decentralised sequencing for rollups, addressing the centralisation of single-sequencer designs, and use restaked ETH to penalise censorship. Proof verification AVS validate zero-knowledge and fraud proofs, adding a layer of security to rollup settlement.
Keeper and Automation Networks: Automation AVS execute liquidations, rebalances and scheduled transactions. These need an economic guarantee that a task runs correctly and on time, which restaked ETH supplies through slashing conditions for missed or incorrect execution.
AVS Reward Mechanics
Each AVS independently determines its reward structure, creating a diverse marketplace of yield opportunities for you. Rewards are usually denominated in the AVS's own token rather than in ETH, which matters more than it sounds: what you actually receive is a token whose price can fall faster than it accrues. The reward flow follows a clear path: the AVS submits rewards, your operator's commission is deducted, and the remainder reaches you through a weekly Merkle distribution root — with a further week's delay before it becomes claimable. Operators set their own rate anywhere from 0% to 100% and may vary it per AVS; the default split is 10% to the operator.
Your reward rate varies across AVS with the restaked ETH allocated to each (more capital means the same rewards split further), the AVS's own revenue, and its maturity. Established AVS such as EigenDA tend to pay less per ETH because they attract more capital; newer ones may pay more to bootstrap security.
An operator validating several AVS stacks rewards from each, so your capital earns from multiple services — the primary attraction of the design. Each additional AVS also adds a slashing vector, which is the trade-off you are actually managing when you pick an operator.
EigenLayer publishes no protocol-level restaking APR, and any figure you see quoted is somebody's estimate rather than a quoted rate. The bulk of what a restaker earns today is Programmatic Incentives: newly minted EIGEN, claimable weekly. The only emission rate the Eigen Foundation has published is the first-year schedule that began on 15 August 2024 — 4% of the initial supply, 3% of it to ETH and LST stakers and their operators. That year has passed with no successor rate published, so no current figure can be quoted honestly. AVS-paid rewards sit on top, and the premium over simply holding a liquid staking token is now small.
How to Restake on EigenLayer
Restaking on EigenLayer involves several steps depending on whether you choose the native or LST path. Both paths ultimately result in your ETH being delegated to an operator who validates AVS on your behalf. Here we walk through both processes with the practical details you need to get started.
Native Restaking Step-by-Step
Native restaking is designed for users who already run or plan to run an Ethereum validator (32 ETH minimum). The process requires modifying your validator's withdrawal credentials to point to an EigenPod smart contract:
- Step 1: Deploy an EigenPod. Connect your wallet to the EigenLayer app and create an EigenPod. This deploys a smart contract that will serve as your validator's withdrawal address. Each wallet address can have one EigenPod.
- Step 2: Set withdrawal credentials. Point your Ethereum validator's withdrawal credentials to your EigenPod address. For new validators, set this during the deposit process. For existing validators, you need to submit a BLS-to-execution-layer withdrawal credential change message.
- Step 3: Verify your validator. Submit a beacon chain state proof to EigenLayer to verify your validator's balance. This proof confirms the amount of ETH available for restaking and is required before you can delegate.
- Step 4: Delegate to an operator. Choose an operator from the EigenLayer operator registry and delegate your restaked balance. Review the operator's AVS portfolio, commission rates, and track record before delegating.
Native restaking provides the highest capital efficiency and avoids LST protocol risk, but requires validator infrastructure knowledge and a minimum of 32 ETH. The withdrawal credential change is irreversible for the lifetime of the validator, so ensure you understand the implications before proceeding.
LST Restaking Step-by-Step
LST restaking is the simpler path, accessible to anyone holding supported liquid staking tokens. The process involves depositing LSTs into EigenLayer's strategy contracts:
- Step 1: Acquire an LST. If you don't already hold a supported LST, stake ETH through Lido (stETH), Rocket Pool (rETH), Coinbase (cbETH), or another supported protocol. Alternatively, purchase LSTs on a DEX.
- Step 2: Approve and deposit. Connect your wallet to the EigenLayer app, approve the LST token for the StrategyManager contract, and deposit your desired amount. The deposit caps that applied during the 2023-24 launch phase no longer apply.
- Step 3: Delegate to an operator. Select an operator and delegate your deposited LSTs. The delegation process is identical to native restaking — choose based on AVS portfolio, commission, and track record.
LST restaking has no minimum deposit requirement beyond gas costs, making it accessible for smaller positions. Your LSTs continue accruing staking rewards whilst deposited in EigenLayer, so you earn both base staking yield and AVS restaking rewards simultaneously. Withdrawals sit in an escrow of WITHDRAWAL_DELAY = 100,800 blocks, roughly 14 days, so that any pending slashing can be applied to the amount before it leaves.
Both paths delegate your stake to the same operator marketplace and earn from the same AVS ecosystem, so the choice turns on entry mechanism and risk layers rather than economics. Native restaking suits you if you already hold 32 ETH or more and run validator infrastructure, because it removes the LST fee layer and one contract dependency. LST restaking suits smaller positions and anyone who would rather hold a transferable token.
Plan your exit before you enter. Undelegating from your operator automatically queues the withdrawal, which then sits in a single escrow of WITHDRAWAL_DELAY = 100,800 blocks, roughly 14 days, before it can be finalised. Rewards stop accruing the moment it is queued, and any pending slashing is applied to the amount before it leaves. If you are restaking an LST, that protocol's own redemption queue sits behind the escrow, so budget for longer than a fortnight.
Yield Analysis and Projections
Understanding EigenLayer's yield structure requires you to separate the components that contribute to your total returns and evaluate the sustainability of each. Your restaking yield is not a single number — it is a composite of base staking rewards, AVS-specific rewards, and potential EIGEN token incentives, each with different risk profiles and sustainability characteristics.
Current Restaking Yields
As of August 2026, your EigenLayer restaking yields break down into the following components:
- Base Ethereum staking yield: around 2.2% APR net, the rate Lido displayed on 4 August 2026, which you earn regardless of EigenLayer participation. That figure is net of Lido's 10% protocol fee, so a solo validator on the native path keeps somewhat more. It has compressed sharply, from roughly 2.6% in early 2026.
- Programmatic Incentives: newly minted EIGEN paid weekly. The Eigen Foundation published a first-year schedule beginning 15 August 2024 — 4% of the initial supply, 3% of it to ETH and LST stakers and their operators — and no successor rate since, so there is no current published number to quote.
- AVS-paid rewards: variable and denominated in each AVS's own token. There is no published aggregate, and with around 20 AVS running live Operator Sets the pool is thin. Rewards stop accruing the moment you queue a withdrawal.
Add those together and the honest answer for August 2026 is that the increment is small against a base staking yield of roughly 2.2% — not the several points that circulated through 2024 and 2025. Realised net returns on the major liquid restaking tokens now run at roughly 2.2% to 2.9% APR, barely distinguishable from plain liquid staking. The dollar value is also price-dependent in a way base staking is not, because you are paid in EIGEN and AVS tokens rather than ETH. If somebody quotes you a firm restaking APR, they are either extrapolating from one AVS or pricing those tokens optimistically.
Your yield also fluctuates with AVS demand cycles: EigenDA rewards rise with rollup activity, oracle AVS rewards track on-chain trading volume. Plan against average conditions rather than peaks, and treat any restaking yield as variable rather than as a rate.
Yield Sustainability Assessment
The design intent is that your restaking yield rests on organic demand — AVS pay for security because they need it, not because of speculative incentives. The reality in 2026 is that the larger and more reliable component is EIGEN emissions on a published schedule, and that schedule is explicitly subject to governance change. Judge the sustainability of the AVS-paid portion separately from the emission-paid portion, because only one of them is evidence of real demand.
Several factors compress your yield further. As restaked capital grows, the same AVS rewards spread across a larger base. The emission component is designed to taper. And because AVS rewards arrive as AVS tokens, the tokenomics of each service feed directly into your realised return: a falling token price cuts what you actually receive even when the nominal reward stream is unchanged. We do not publish a forward estimate of a sustainable restaking APR: with the entire restaked economy at roughly $5 billion and only a handful of AVS paying, any such number would be an extrapolation dressed as a forecast.
The key metric you should watch is AVS revenue growth relative to restaked capital. If AVS revenue grows faster than restaked capital, your yields improve; if the reverse, they compress. Through 2026 both have been shrinking together — restaked value fell from roughly $22 billion in August 2025 to about $5 billion a year later — which tells you the demand side has not yet grown into the security on offer.
Security Assessment and Risks
Restaking fundamentally changes the risk profile of your staked ETH. Whilst simple Ethereum staking exposes you to a single set of slashing conditions (Ethereum consensus rules), restaking adds additional slashing vectors from every AVS your operator validates. This layered risk structure is the most important consideration for you when evaluating EigenLayer, and you must understand it thoroughly before committing your capital.
Slashing Mechanics in Detail
EigenLayer implements a two-tier slashing system that separates Ethereum consensus slashing from AVS-specific slashing:
Ethereum consensus slashing applies to you as a native restaker if your validator violates beacon chain rules (double voting, surround voting). Since the Pectra upgrade the initial penalty is 1/4096 of your validator's effective balance — roughly 0.008 ETH on a 32 ETH validator, against about 1 ETH before Pectra — although the correlation penalty can still reach the entire balance in a mass slashing event. This risk exists regardless of your EigenLayer participation and is well-understood by the staking community.
AVS slashing is the novel risk that restaking introduces for you. Each AVS defines its own slashing conditions — specific behaviours that, if detected, result in a portion of the stake your operator allocated to that service being burned or, in a redistributable Operator Set, paid to a recipient the AVS named in advance. For example, an oracle AVS might slash operators who submit incorrect price data, whilst a bridge AVS might slash operators who sign invalid cross-chain messages. The slashing amount and conditions vary by AVS, and if your operator validates multiple AVS, you face cumulative slashing exposure from all of them.
Two details here are easy to miss and both cut against you. EigenLayer's documentation states that slashing "does not have to be objectively attributable (that is, provable onchain)", so an AVS is not obliged to prove the fault on-chain. And since ELIP-006 an AVS may create a redistributable Operator Set, in which seized stake is paid to an address the AVS fixed in advance rather than burned — EigenLayer itself warns this creates "a larger incentive to slash user funds". Fourteen such Operator Sets existed on mainnet by August 2026.
Be precise about what actually protects you, because this point is widely misreported. There is no protocol-level slashing veto: EigenLayer's own developer documentation states that "no vetoes are provided by the EigenLayer protocol", and a veto committee is an optional pattern an individual AVS may choose to build for itself. What the protocol gives you instead is time. Newly allocated stake only becomes slashable after an allocation delay; deallocation takes 100,800 blocks, about 14 days; and since ELIP-016 shipped in June 2026 there is a further resolution delay of 50,400 blocks, roughly seven days, between an AVS calling a slash and the shares being burned or redistributed. Those windows let you undelegate. They do not give you an appeal.
Before you delegate, check one thing specifically: whether your chosen operator has joined any redistributable Operator Set. The difference matters to you. In a standard set, slashed stake is burned and nobody profits from taking it. In a redistributable set, an address named in advance receives it — which changes the incentive facing whoever controls the slashing decision. Is the extra reward on those sets worth that? For most stakers it is not.
Slashing is live, and it has fired — but barely. An on-chain scan of the AllocationManager contract to 4 August 2026 finds 15 OperatorSlashed events in total, spanning September 2025 to June 2026 and originating from just three AVS addresses; most carry free-text descriptions that read as tests or marketing rather than genuine service faults, and EigenLayer's burn address still holds a zero balance of every major restaked LST. Read that carefully: the mechanism works, but it has never been exercised in anger against a large third-party operator. That is a very different statement from "restaking has been proven safe", and the operators carrying the largest delegations have simply not yet been tested by a real dispute.
Smart Contract and Systemic Risks
Beyond slashing, EigenLayer introduces smart contract risk at multiple layers that you must consider. The core EigenLayer contracts (StrategyManager, DelegationManager, EigenPodManager) manage billions in restaked assets and represent a significant attack surface. A vulnerability in any of these contracts could potentially affect your restaked capital. This is the single largest risk factor you face in the protocol.
Your risk compounds as an LST restaker, because you carry contract risk from the LST protocol and EigenLayer at once, and a third layer again if you hold a liquid restaking token. Each layer multiplies the chance of meeting a vulnerability even when every individual protocol is well audited. This is not abstract: in April 2026 Kelp DAO suffered a $292 million bridge exploit that minted roughly 116,500 unbacked rsETH. It was a bridge configuration failure at the LRT layer that never touched EigenLayer's core contracts — and holders were made whole only because a coalition of DeFi protocols recapitalised the shortfall, not because any mechanism guaranteed it.
Systemic risk sits above all of this. Shared security creates interdependence: a major slashing event on one AVS could drive outflows that reduce the security available to every other AVS, and a large operator slashed across several sets at once would hit its delegators hard. These tail risks are hard to quantify but worth acknowledging when you size the position.
Governance risk is the last layer. Safety-delay parameters, the reward emission schedule and the rules governing redistributable slashing are all governed decisions that can materially change your position after you have taken it — an Incentive Council was added to the contracts in March 2026 with a hand on the reward schedule. Because the emission schedule is the larger part of what you are paid, a governance change there hits your return directly rather than at the margin. Follow the ELIP process if you hold a meaningful position.
Audit History and Bug Bounties
Before you commit capital, review the audit reports EigenLayer publishes in its contracts repository. Four firms account for the bulk of the work:
- Sigma Prime — the deepest coverage, from the M1 mainnet release in May 2023 through rewards v2, permissionless strategies and the v1.0.0 slashing release
- Certora — formal verification work, including the slashing release, the multichain contracts and the redistribution slash delay in May 2026
- Cantina — the M2 mainnet release in April 2024 and the v1.0.0 slashing release in March 2025
- Consensys Diligence — the original M1 mainnet contracts in March 2023
Note what that list does not contain: Trail of Bits and ABDK Consulting are frequently attributed to EigenLayer in third-party write-ups, and neither appears in the published audit set. The slashing release itself carries three independent reviews, which is the right level of scrutiny for the component that can take your money. The protocol also runs a bug bounty through Immunefi. The combination of repeated audits and over two years of mainnet operation with no protocol-level exploit should give you reasonable confidence in the core contracts — though you must accept that no smart contract system can be considered completely risk-free.
For a comprehensive analysis of slashing mechanics, depeg scenarios, and smart contract layering risks across the liquid staking ecosystem, see our dedicated liquid staking risks analysis.
Operator Selection Guide
Choosing the right operator is the most consequential decision you face here, because your operator alone determines which Operator Sets your stake is allocated to and therefore what can slash it. Note the constraint before you plan around it: delegation is all-or-nothing to a single operator per address, so "diversifying across operators" means running separate addresses, not splitting one balance.
Track record and reputation. Prioritise operators whose history in Ethereum infrastructure you can verify — validator operators and staking services that have run reliable systems for years. High uptime on Ethereum validators is the best available proxy for reliable AVS validation. Check whether they have been slashed before and how they handled it.
AVS portfolio and allocations. Review which Operator Sets the operator has joined and how much Unique Stake they have allocated to each, because that allocation is the ceiling on what any single AVS can take from you. More sets means more potential reward and more cumulative slashing exposure. Allocations change, so this is a position to monitor rather than a decision to make once.
Commission structure. Operators set their own rate — the protocol permits anything from 0% to 100%, with 10% as the default split — and may vary it per AVS. Lower is better for you, but a rate far below the field can signal an operator economising on infrastructure. Compare within a peer group running similar Operator Sets.
Infrastructure quality and size. Before you commit, look for published detail on geographic distribution, redundancy, monitoring and incident response, plus uptime statistics. On size, very large operators bring economies of scale but concentrate ecosystem risk, while very small ones may lack the resources for professional infrastructure; mid-sized operators often balance the two.
Ask for specifics rather than assurances. Does the operator publish uptime statistics you can check against an independent dashboard? Do they run in more than one region with automated failover, so that a single data-centre outage does not turn into missed validation tasks? Have they written up a past incident, and does that write-up read like an engineering post-mortem or like a press release? You should treat an operator who will not answer these in public as a worse risk than one who admits to a bad week.
The EigenLayer app's operator registry is your practical starting point, listing delegated stake, Operator Sets, commission and performance. Use it to shortlist, then research each candidate's background and community reputation before you delegate.
EigenLayer vs Alternatives
EigenLayer is the protocol that defined restaking, but it is not your only option for seeking additional yield on staked ETH. Understanding how EigenLayer compares to alternatives helps you contextualise its strengths and limitations for your situation.
EigenLayer vs Liquid Restaking Tokens (LRTs). Protocols like Ether.fi (eETH), Kelp DAO (rsETH), and Puffer Finance (pufETH) are not competitors to EigenLayer — they are built on top of it. These protocols deposit ETH into EigenLayer on your behalf and issue liquid tokens representing your restaked position. They add convenience and DeFi composability for you but also add an additional smart contract layer. If you want direct control over operator selection and AVS exposure, restake directly on EigenLayer.
One caveat about LRTs matters more than any other in 2026: Ether.fi is unwinding its restaking exposure entirely. Its own security documentation states that under 1% of its assets remained restaked with EigenLayer as of August 2026, down from about half in early 2026, targeting zero by the third quarter of 2026 and removal of EigenPod withdrawal credentials by the fourth. Buying an LRT is therefore no longer a dependable way to get restaking exposure — check what the specific token is actually restaked into before you assume it is.
EigenLayer vs Symbiotic. Symbiotic is an alternative restaking protocol built around modular vaults, with fewer protocol-level restrictions on collateral types and slashing parameters. The collateral distinction that used to separate the two has narrowed: EigenLayer now supports permissionless token strategies, so any ERC-20 can be added as restaking collateral there too. What remains is a difference of design and history rather than one we can size: we have no current restaked-value figure for Symbiotic to set against EigenLayer's roughly $5 billion, down from $22 billion a year earlier. EigenLayer has run operator delegation on mainnet since April 2024, and that record rather than scale is the case for treating it as the more conservative choice.
EigenLayer vs simple staking. The most relevant comparison for you may be whether restaking is worth the additional complexity and risk compared to simple Ethereum staking (via Lido, Rocket Pool, or solo validation). Simple staking pays around 2.2% APR as of August 2026 with well-understood risks and a redemption path that has never been impaired. Restaking adds a token-denominated increment that realised returns put in tenths of a percentage point, and introduces AVS slashing risk, additional smart contract risk and a 14-day exit escrow. On those numbers, restaking is only worth it if you have a specific view on the AVS you are backing.
For a detailed side-by-side comparison of EigenLayer with Ether.fi and Kelp DAO, including TVL, yields, risk profiles, and DeFi integration, see our restaking comparison.
Pros and Cons
Pros:
- Additional yield on already-staked ETH without committing additional capital
- Largest restaking ecosystem, with around 20 AVS running live Operator Sets on mainnet
- Non-custodial delegation model — operators never take custody of your assets
- Supports both native restaking (solo validators) and permissionless ERC-20 strategies
- Repeated audits from Sigma Prime, Certora, Cantina and Consensys Diligence, with three independent reviews of the slashing release alone
- Over two years of mainnet operation with no protocol-level exploit
- Safety delays are explicit and on-chain, so you can see exactly how long you are locked
Cons:
- Additional slashing risk from AVS — each AVS adds a new vector for potential penalties
- Smart contract risk on top of base staking risk — multiple contract layers increase attack surface
- Operator dependency — poor operator selection can result in slashing or suboptimal yields
- Operator and AVS concentration — only about 29 operators have ever registered into a slashable Operator Set, and three AVS addresses account for every slashing event on record
- 14-day withdrawal escrow, during which rewards stop accruing, reduces liquidity compared with liquid staking tokens
- Most of the current yield is EIGEN emissions on a governance-adjustable schedule, not AVS revenue
- Complexity barrier — understanding AVS, operators, and slashing requires significant research
- Restaked value has fallen roughly 77% from its August 2025 peak, and the AVS demand side remains thin
- No protocol-level veto on slashing, and slashing need not be objectively provable on-chain
How should you weigh these trade-offs in practice? The slashing risk is mitigated by delay, not by appeal: the seven-day resolution window and the 14-day deallocation delay give you time to undelegate if an operator's allocations change, but nobody at protocol level will reverse a slash on your behalf. The 14-day exit escrow is still shorter than several proof-of-stake networks — Cosmos chains typically require 21 days — though it is twice the figure many older write-ups quote. On the positive side, the non-custodial delegation model means your assets remain in smart contracts you can verify on-chain, not on a corporate balance sheet where counterparty risk dominates.
The obvious escape valve — hold a liquid restaking token instead — has weakened. Ether.fi is winding its restaking exposure down to zero, and Kelp's rsETH bridge was exploited in April 2026, so an LRT now buys you a different risk rather than less of it.
Should you restake on EigenLayer? If you can evaluate operators, understand AVS risk profiles, and manage your positions actively, the balance still works — but the reward for that work is thinner than it was. If you prefer simple, passive yield without ongoing research, plain liquid staking is the better fit. Our 4.3/5 rating reflects sound engineering and no protocol-level exploit, tempered by a shrinking demand side and risk layers you must navigate yourself.
Conclusion
EigenLayer represents a genuine innovation in blockchain security design that you should understand whether or not you choose to restake. By enabling your staked ETH to secure additional services beyond Ethereum consensus, the protocol creates a shared security marketplace where you earn more yield on the same capital, AVS builders access robust security without bootstrapping costs, and the Ethereum ecosystem becomes more capital-efficient for everyone involved.
Should you trust EigenLayer with your staked ETH? On engineering, the answer is largely yes: over two years of mainnet operation, repeated independent audits with three separate reviews of the slashing release, and no protocol-level exploit. On economics, be much more cautious. Restaked value has fallen from roughly $22 billion in August 2025 to about $5 billion a year later, the yield premium over plain staking is now a fraction of a percentage point rather than the several points once advertised, and the larger part of what you are paid arrives as EIGEN emissions on a schedule governance can change. The security machinery is finished; the demand it was built to serve has not arrived at the scale the design assumed.
However, you must honestly assess whether restaking suits your investment style. The additional slashing vectors, smart contract layers, and operational complexity make it unsuitable if you want simple, predictable staking returns. You need to engage actively — choosing your operators carefully, understanding AVS risk profiles, monitoring your positions, and staying informed about ecosystem developments. If you are unwilling or unable to invest this effort, you should consider straightforward liquid staking through Lido or Rocket Pool instead.
If you understand the risks and can manage them, EigenLayer remains the most credible venue for restaking — but treat it as a considered allocation rather than an obvious upgrade to staking. What shipped in 2026 was risk plumbing rather than growth: redistributable slashing under ELIP-006, which lets an AVS pay seized stake to a recipient of its choosing, and the seven-day slash resolution delay under ELIP-016. Our 4.3/5 rating reflects a protocol that delivers on its core technical promise whilst acknowledging that the complexity, the thin AVS demand and the risk layers are real trade-offs you must evaluate for yourself. Whether restaking recovers its earlier scale depends on AVS revenue, and that has not yet turned. To start restaking with the most popular liquid staking protocols, see our Binance staking referral guide and Rocket Pool referral guide.
Sources and References
Frequently Asked Questions
- What is EigenLayer and how does restaking work?
- EigenLayer is a middleware protocol on Ethereum that allows staked ETH (or LSTs like stETH and rETH) to be restaked to secure additional services called Autonomous Verifiable Services (AVS). Restakers delegate their stake to operators who validate AVS tasks, earning extra rewards on top of base Ethereum staking yield. The protocol extends Ethereum's cryptoeconomic security to new applications without requiring them to bootstrap their own validator sets. You can restake either natively (as a solo validator via EigenPods) or by depositing LSTs into EigenLayer's strategy contracts.
- How much yield can I earn restaking on EigenLayer?
- EigenLayer publishes no protocol-level restaking APR, so treat any quoted figure as somebody's estimate. Most of what a restaker receives is Programmatic Incentives — newly minted EIGEN — with AVS-paid rewards on top. The only emission rate the Eigen Foundation has published is the first-year schedule that began on 15 August 2024: 4% of initial supply, 3% of it to ETH and LST stakers and their operators. That year has passed with no successor rate published, so no current number can be quoted honestly. Base Ethereum staking was around 2.2% APR net on Lido on 4 August 2026, and realised returns on the major liquid restaking tokens run at roughly 2.2% to 2.9%, so the restaking premium is small and paid in tokens rather than ETH.
- What are the slashing risks in EigenLayer?
- EigenLayer introduces additional slashing conditions beyond standard Ethereum slashing. Each AVS defines its own criteria, and an AVS may take up to the whole of the stake an operator allocated to its Operator Set. EigenLayer's documentation states that slashing "does not have to be objectively attributable (that is, provable onchain)", and that "no vetoes are provided by the EigenLayer protocol" — a veto committee is an optional pattern each AVS may build for itself, not a protocol backstop. What you get is delay: an allocation delay, a 14-day operator deallocation delay, and a seven-day resolution delay before slashed shares are burned or redistributed. On-chain records to August 2026 show 15 slashing events in total from three AVS addresses, most of which read as tests rather than genuine faults.
- How do I choose an EigenLayer operator?
- Evaluate operators on their track record in Ethereum infrastructure, total delegated stake, which Operator Sets they have joined and how much Unique Stake they allocate to each, their commission, and infrastructure reliability. Commission is not standardised: the protocol permits anything from 0% to 100%, with 10% as the default split, and operators may vary it per AVS. Note that delegation is all-or-nothing to one operator per address, so spreading across operators requires separate addresses rather than splitting a balance. Allocations change over time, so review them periodically rather than treating the choice as one-off.
- Is EigenLayer safe for long-term restaking?
- EigenLayer's published audit set is dominated by Sigma Prime, Certora, Cantina and Consensys Diligence; Trail of Bits and ABDK Consulting are often attributed to the protocol in third-party write-ups and appear nowhere in it. The slashing release alone carries three independent reviews. The protocol has run AVS delegation on mainnet since April 2024 with no protocol-level exploit and runs a bug bounty through Immunefi. However, restaking adds smart contract layers and slashing conditions that do not exist in simple staking, and the economics have deteriorated — restaked value is down roughly 77% from its August 2025 peak. For long-term positions, choose operators conservatively, monitor their Operator Set allocations, and size accordingly.
- What is the difference between native restaking and LST restaking?
- Native restaking is for solo validators who point their withdrawal credentials to an EigenPod smart contract, allowing their 32 ETH validator balance to serve as restaking collateral. LST restaking allows holders of liquid staking tokens (stETH, rETH, cbETH) to deposit them into EigenLayer's strategy contracts. Native restaking offers higher capital efficiency and avoids LST protocol risk but requires validator infrastructure and 32 ETH minimum. LST restaking is simpler, has no minimum beyond gas costs, and is accessible to any LST holder.
- Can I withdraw my restaked ETH at any time?
- Not immediately. Undelegating automatically queues the withdrawal, which then sits in a single escrow of WITHDRAWAL_DELAY = 100,800 blocks, roughly 14 days — not the 7 days often quoted. The delay exists so that any pending slashing can be applied to your amount before it leaves the protocol, which means the figure you eventually receive can be lower than the one you queued. Rewards stop accruing the moment you queue the withdrawal. A liquid restaking token can be sold on a decentralised exchange without waiting, but that is a different risk rather than less of it: Ether.fi is winding its EigenLayer exposure down, targeting zero by the third quarter of 2026, and Kelp's rsETH bridge was exploited in April 2026.
Financial Disclaimer
This content is not financial advice. All information provided is for educational purposes only. Cryptocurrency investments carry significant investment risk, and past performance does not guarantee future results. Always do your own research and consult a qualified financial advisor before making investment decisions.