What Is a Liquid Restaking Token (LRT): Complete DeFi Guide (2026)

If you have been anywhere near Ethereum DeFi in the last two years, you have probably heard the acronym LRT thrown around in farming threads, yield dashboards, and Discord servers. Liquid restaking tokens exploded from a niche EigenLayer experiment into one of the largest categories in decentralized finance, with tens of billions of dollars locked across protocols like ether.fi, Renzo, KelpDAO, and Puffer Finance. If you blink, the leaderboard reshuffles.

An LRT, at its simplest, is a tokenized receipt for ETH that is simultaneously staked on Ethereum to secure the base layer and restaked on EigenLayer to secure additional services called AVSs. That single token captures three or four yield streams at once: base ETH staking, AVS rewards, native LRT points, and (until recently) EigenLayer restaking points. The catch is that those stacked yields come with stacked risks, and most articles in the first page of Google still confuse LST with LRT, omit half the protocols that matter, or skip the math on correlated slashing entirely.

This guide fixes that. We walk through what an LRT actually is, the precise difference between a liquid staking token and a liquid restaking token, the eight protocols currently fighting for LRT market share in 2026, how the yield stacks (and how much of it is real), how LRTs plug into Pendle, Aave, and Curve to form an ecosystem people now call LRT-fi, what correlated slashing means in mathematical terms, and how the post-EIGEN TGE era has changed the points farming game forever. By the end you should be able to look at any LRT dashboard and know exactly what you are buying.

What Is a Liquid Restaking Token (LRT)?

A liquid restaking token is an ERC-20 token issued by a protocol that takes your ETH (or your existing liquid staking token like stETH), stakes it on Ethereum so it earns base validator yield, and then restakes the same staked ETH on EigenLayer so it can also secure third-party services called Actively Validated Services (AVSs). In return, you receive a fungible token that represents your share of the underlying restaked ETH plus all the rewards being generated on top of it.

You can think of an LRT as a stacked receipt. The first layer of the stack is ordinary Ethereum staking, currently paying around three to four percent annualized. The second layer is restaking, which delivers extra rewards in the native tokens of the AVSs your LRT is securing, plus EigenLayer ecosystem incentives. The third layer is usually a points program from the LRT protocol itself, which converts into an airdrop later. The fourth layer is DeFi composability, where you take the LRT and deploy it into Aave, Pendle, Curve, or a yield aggregator to earn even more on top.

The critical mental model is this: an LRT is not a brand-new asset class. It is a liquid staking token plus restaking exposure, wrapped into a single transferable ERC-20. If you understand how liquid staking works and you understand how EigenLayer restaking works, you already understand 90% of LRTs. The remaining 10% is the protocol-specific design choices and risk profile of each issuer.

LST vs LRT: The Key Distinction

This is where most beginner articles fail. They use LST and LRT interchangeably or describe LRTs as "the new generation of LSTs," which is misleading. The relationship is more like LSTs are the ingredient, LRTs are the dish.

A liquid staking token, or LST, is a tokenized claim on staked ETH. You give a protocol like Lido or Rocket Pool one ETH, they run a validator with it, and they issue you a token like stETH or rETH in return. That token earns the base Ethereum staking yield, currently 3 to 4 percent APY, and you can trade it, lend it, or use it as collateral while it keeps accruing. The LST exists for one job: make staked ETH liquid.

A liquid restaking token, or LRT, is one step higher in the stack. The LRT protocol takes either your raw ETH or your existing LST, stakes it on Ethereum if needed, and then restakes that staked ETH on EigenLayer. The restaking step delegates your ETH to AVSs that pay extra rewards in exchange for using your stake as economic security. The LRT you receive in return earns the base staking yield AND the restaking rewards, all in one wrapper.

To make it concrete: stETH is an LST. It earns roughly 3.5% APY from Ethereum validator rewards. eETH from ether.fi is an LRT. It earns the same 3.5% base staking yield plus restaking rewards from EigenLayer AVSs plus ether.fi loyalty points plus historically EigenLayer points. Every LRT has an LST inside it, but not every LST is an LRT.

How LRT Yield Stacks: The Four Layers

Understanding LRT yield requires breaking it into the components, because the headline APY numbers protocols advertise often blend real cash flows with speculative future airdrops. Let us unstack each layer carefully.

Layer 1: Base ETH Staking. This is the bedrock. Every ETH locked into an LRT is running on an Ethereum validator somewhere. Validators earn priority fees, MEV tips, and the protocol-level issuance reward for proposing and attesting blocks. In 2026, with around 35 million ETH staked, this yield sits between 2.8% and 3.5% APY depending on network activity. This portion is paid in real ETH and is the most reliable component.

Layer 2: AVS Rewards. Restaked ETH is delegated to AVSs. These are services like data availability layers (EigenDA), bridges, oracle networks, coprocessors, and rollup security providers. Each AVS pays operators in their native token (or in ETH) for providing economic security. As of 2026, AVS rewards are real but uneven. Some AVSs pay generously to attract early restakers, others pay almost nothing. Averaged across a diversified LRT, this layer typically adds 1% to 3% APY in nominal value, though much of it is paid in volatile altcoins.

Layer 3: LRT Native Points. Every major LRT runs a loyalty program. ether.fi has Loyalty Points, Renzo has ezPoints, KelpDAO has Kelp Miles. These accrue daily based on your LRT balance and historically converted into airdrops worth anywhere from 5% to 25% of deposit value. This is the most variable layer. The first wave of LRT airdrops in early 2024 was extraordinarily lucrative. Subsequent rounds have been smaller as the market matured and farming became more efficient.

Layer 4: EigenLayer Points and Other Programs. Restaked ETH historically also earned EigenLayer restaking points, which converted into the EIGEN token at the TGE in 2024. After the TGE, EigenLayer transitioned to a season-based incentive program. Points are still earned, but the conversion rate to tokens is now lower and more transparent. Some LRTs also farm secondary points from AVSs that have their own pre-token incentive programs.

Top 8 LRT Protocols in 2026

The LRT space consolidated around eight major issuers, each with a different design philosophy, validator strategy, and approach to AVS selection. Here is the lineup, ranked roughly by TVL and ecosystem reach.

ether.fi (eETH) Deep Dive: The TVL Leader

ether.fi spent most of 2024 and 2025 sitting at the top of the LRT TVL charts and has held that position into 2026. The protocol's core innovation is non-custodial restaking. When you deposit ETH, ether.fi assigns it to a validator, but the validator key is held by the node operator rather than the protocol or a third-party custodian. This is meaningfully different from most LSTs and LRTs, where the staking protocol or a federated set of operators holds the keys.

You deposit ETH and receive eETH, which rebases daily to reflect staking and restaking rewards. eETH can be wrapped into weETH, the non-rebasing version that integrates more cleanly with DeFi protocols that do not support rebase tokens. Almost all DeFi integrations actually use weETH. eETH and weETH together represent the largest LRT supply in the ecosystem.

The ether.fi loyalty program (Loyalty Points) was one of the most lucrative campaigns of the 2024 cycle. After multiple seasons and the ETHFI token launch, the protocol moved to a structured points program with clearer conversion paths. ether.fi also expanded into Liquid, a vault product that automatically deploys LRT yield strategies, and Cash, a debit card backed by ETH collateral that ties consumer finance into the protocol.

Renzo (ezETH): The Cross-Chain LRT

Renzo built its market share by being everywhere. While ether.fi focused on mainnet depth, Renzo aggressively deployed ezETH to Arbitrum, Base, BNB Chain, Linea, Mode, and Blast. This cross-chain footprint made ezETH the default LRT for users on L2s who did not want to bridge back to mainnet to participate in restaking.

Renzo accepts ETH directly, as well as wBETH and stETH, which gives flexibility to users already holding LSTs. The protocol routes ETH to a curated set of operators, restakes the result through EigenLayer, and selects AVSs based on a risk-weighted framework. The Renzo team has been more public than many about its operator and AVS selection criteria, which is helpful for due diligence.

One incident worth knowing: in April 2024, ezETH suffered a brief depeg when the REZ token airdrop was announced with terms that disappointed some farmers, triggering a cascade of liquidations across leveraged positions. The token traded as low as 0.85 ETH for several hours before recovering. This event is now the textbook example of why LRT liquidity matters as much as LRT yield.

KelpDAO (rsETH): The Multi-LST Strategy

KelpDAO took a different angle: instead of starting from ETH and creating its own validator network, Kelp accepts existing LSTs and packages them into a single restaked token. Deposit stETH, ETHx, sfrxETH, ankrETH, or mETH, and you receive rsETH. Kelp then restakes the underlying through EigenLayer.

This design has two implications. First, you gain diversification across the underlying LSTs, which matters because each LST has its own validator set and slashing risk. Second, you inherit the price dynamics of multiple LSTs at once. rsETH tracks ETH closely but with a baked-in basket exposure that smooths out idiosyncratic depegs in any single underlying LST.

Kelp Miles, the loyalty program, converted into the KEP token (Kelp Earned Points) and later into governance airdrops. The KelpDAO governance model is more structured than some competitors, with on-chain voting on AVS allocation and operator inclusion. For users who care about decentralization beyond the yield numbers, Kelp's governance is a real differentiator.

Puffer (pufETH): Anti-Slashing Tech

Puffer Finance's pitch is hardware-enforced slashing protection. Most LRT protocols rely on operator behavior and software-level safeguards to avoid slashing. Puffer goes further by requiring node operators to run Secure-Signer, a piece of hardware-enforced signing software that physically prevents the validator from signing two conflicting attestations or proposing two conflicting blocks. The signing key is locked inside hardware that the operator cannot override.

The practical effect is that the most common cause of slashing, double signing, becomes physically impossible on Puffer's network. Deposit ETH and receive pufETH, which represents your share of the slashing-protected validator set plus the EigenLayer restaking position.

Puffer also pioneered a permissionless validator onboarding model. Anyone can spin up a Puffer validator with a smaller ETH bond than running a solo validator would require, because the Secure-Signer hardware reduces the slashing risk to the protocol. This made Puffer a favorite of solo stakers who wanted to participate in restaking without committing 32 ETH per validator.

Swell, Mellow, Bedrock, and Eigenpie

The remaining four protocols round out the LRT ecosystem with distinct angles.

Swell (rswETH) evolved from the Swell Network's earlier LST product (swETH) into a full restaking offering. Swell's calling card is its integrated L2 rollup, which is itself secured by restaked ETH from rswETH holders. This creates a closed-loop ecosystem where holding rswETH gives you exposure to both the AVS rewards and the success of Swell's own rollup product.

Mellow LRT is not a single LRT, it is a factory. Mellow provides modular infrastructure for anyone to launch their own LRT vault with custom risk parameters, AVS selection, and operator choices. Multiple LRTs have been deployed using Mellow, including LRTs curated by specific funds and DAOs. This is the "white-label LRT" approach.

Bedrock (uniETH) comes from the RockX team, which has institutional staking infrastructure dating back to the Beacon Chain launch. Bedrock leans into compliance-friendly design and offers restaking products that go beyond ETH, including BTC and IOTX. uniETH is the ETH product and follows a more conservative AVS selection policy.

Eigenpie (meTH) sits inside the Magpie ecosystem, which also produces yield-boost products on Penpie (a Pendle booster) and other yield aggregators. Eigenpie issues isolated LRTs per underlying LST: deposit stETH and get mstETH, deposit ETHx and get methx, and so on. The isolation makes it easier for sophisticated users to manage exposure to specific LSTs while still capturing the restaking layer.

LRT-fi: DeFi Composability Map

The reason LRTs grew so fast is not just the yield. It is the composability. Every major LRT is integrated into a network of DeFi protocols that allow you to leverage, swap, separate, or aggregate the yield. This network has its own name now: LRT-fi.

Pendle is arguably the most important venue in LRT-fi. By splitting an LRT into a Principal Token (PT) and a Yield Token (YT), Pendle lets users either lock in a fixed yield by buying PT at a discount, or buy concentrated exposure to all future yield (including points) by buying YT. During the heaviest points farming periods of 2024, YT positions on weETH-Pendle were the single highest-yielding strategy in DeFi for a brief window.

Aave and Spark accepting LRTs as collateral was the second major composability unlock. Once you can borrow ETH against weETH, you can recursively deposit and borrow to lever up your LRT exposure. A typical loop might be: deposit weETH, borrow ETH at 80% LTV, swap ETH back to weETH, redeposit, repeat. With 5x leverage, the underlying LRT yield is multiplied by five, but so is the depeg risk and the liquidation risk.

Pendle PT/YT on LRTs: Yield Separation Explained

Pendle deserves its own section because it is the most misunderstood part of LRT-fi. Pendle is a protocol that splits any yield-bearing asset into two tokens: a Principal Token (PT) that redeems for 1 unit of the underlying at maturity, and a Yield Token (YT) that captures all yield generated by the underlying until maturity.

For an LRT like weETH, this means: buy a weETH-PT at a discount (say 0.95 ETH worth), and at maturity it converts back to 1 weETH worth of value. You earn the spread, which behaves like a fixed-yield bond. Or buy weETH-YT, which is much cheaper because it has no principal, but you capture every staking reward, restaking reward, and loyalty point earned until maturity. YT positions are the maximum-leverage way to farm points.

During the peak ether.fi farming season, weETH-YT effectively gave users 30 to 50 times leveraged points exposure on the capital they spent buying it. The trade-off is that if the points are worth less than expected at TGE, YT goes to zero. A whole subset of LRT-fi traders specialize in modeling expected airdrop value against current YT prices.

LRTs as Aave/Spark Collateral

Most LRTs are now accepted as collateral on Aave V3 and Spark. The mechanic is simple: deposit weETH or ezETH, and the protocol values it at a discount to ETH (typically 92 to 95 percent) and lets you borrow ETH, USDC, or other assets against it. The collateral keeps earning the LRT yield while you take out a loan.

The killer strategy is the looping trade. Deposit 1 weETH, borrow 0.8 ETH, swap back to 0.78 weETH (accounting for slippage), redeposit. After several iterations you might end up with effective exposure of 4 to 5 weETH for every 1 weETH of capital. The leveraged yield is impressive on paper. The risk profile, however, is wrapped around three failure modes: weETH depegging (collateral value crashes), borrow rates spiking (your loan becomes expensive), or oracle issues (liquidation cascades).

The April 2024 ezETH depeg flushed roughly $60 million in leveraged positions in a few hours, mostly users who were running aggressive Aave loops. This is the canonical lesson that LRT-on-Aave leverage is not free yield. It is a directional bet on the LRT staying pegged.

Correlated Slashing Risk: The Math

Here is the topic almost every other guide ignores. When you restake ETH on EigenLayer, you delegate the same ETH to multiple AVSs simultaneously. If any single AVS slashes your operator, you lose ETH. The crucial question is: how independent are the slashing risks across AVSs?

EigenLayer's design includes Unique Stake mechanisms that partially address this by allowing AVSs to require dedicated stake that cannot be double-counted across services. Several LRTs also implement operator diversification policies that cap how much stake goes to any single operator. Reading the LRT's restaking policy document, if one exists, is the only way to know how exposed you actually are.

The other layer of risk is the AVS itself. Not every AVS implements slashing. Some are still in "rewards-only" mode where you earn AVS tokens but cannot be slashed. As more AVSs activate slashing, restaking moves from a free-money phase into a real-risk phase. Users who joined LRTs in 2024 for points may not have realized that the underlying restaking is becoming meaningfully more dangerous over time.

The EIGEN TGE Era and Post-Airdrop Yield Reality

The launch of the EIGEN token in 2024 was a turning point for the LRT ecosystem. Before TGE, LRT users were accumulating EigenLayer points with no concrete pricing. The expected airdrop value was a speculative number that could only be inferred from secondary points markets and protocol valuations.

After TGE, the value of past points crystallized, and the EIGEN token became a tradable asset. The market priced restaking yield much more rationally. Pure points farming, where users would deposit billions of ETH chasing speculative airdrops, dropped sharply as a strategy. The remaining yield had to justify itself on AVS rewards plus real protocol cash flows.

The post-TGE reality is that LRT yield is settling into a more sustainable range. Headline APYs of 15% or 20% from the points-farming era have come down to 4% to 7% all-in. The 4% base comes from ETH staking, and the additional 0% to 3% comes from real AVS rewards. Native LRT loyalty programs still add a speculative layer, but the magnitude is smaller. The result is a more mature market with clearer risk-adjusted returns.

Several LRTs adapted by launching their own tokens. ETHFI from ether.fi, REZ from Renzo, and KEP from KelpDAO all created new yield surfaces tied to their token's restaking gauges. Holders of LRT tokens can sometimes restake or stake the governance tokens for additional yield, although this is third-order leverage and carries its own risks.

How to Choose an LRT: A Decision Framework

With eight major protocols and dozens of smaller ones, picking an LRT is no longer trivial. Here is the framework experienced users apply.

Step 1: Define your time horizon. Are you holding for one month or one year? Short-term holders should optimize for liquidity and DeFi integration depth (ether.fi, Renzo). Long-term holders should prioritize operator quality and slashing protection (Puffer, Bedrock).

Step 2: Check secondary market liquidity. An LRT is only as good as your exit. Look at the LRT/ETH Curve or Balancer pool depth. If you cannot exit $1 million at less than 0.5% slippage, you do not own a liquid asset. The 2024 ezETH event proved that thin liquidity transforms small redemption delays into large depegs.

Step 3: Read the AVS policy. The LRT's documentation should specify which AVSs it delegates to and how aggressively. More AVSs equals more yield equals more slashing exposure. A conservative LRT that delegates to two well-vetted AVSs is a different product from an aggressive LRT delegating to a dozen.

Step 4: Evaluate the token and points status. Has the protocol launched its token already? If so, the points-airdrop layer is gone or reduced. If not, you may still capture an upcoming airdrop, but the timing is uncertain. Adjust your expected yield accordingly.

Step 5: Confirm DeFi composability matches your strategy. If you want to loop on Aave, your LRT must be accepted as collateral on Aave. If you want to farm Pendle YT, your LRT must have an active Pendle market with adequate liquidity. Not every LRT is integrated everywhere.

For a deeper background on how restaking fits with liquid staking, our complete restaking guide and our EigenLayer protocol guide walk through the underlying mechanics. If you want to understand the validator-level layer that pays the base staking yield in every LRT, our PoS validator guide is the place to start. And if you are considering the simpler LST path without the restaking layer, our staking pool guide covers that side.

Risks: What Can Actually Go Wrong

LRTs have a longer risk surface than almost any other DeFi product, because they stack three protocols deep. Here are the failure modes in priority order.

Slashing cascade. Discussed above in detail. The interaction between AVS slashing and operator concentration is the most underappreciated risk. A single high-correlation slashing event could affect multiple AVSs on the same operator simultaneously, and the loss flows through to LRT holders.

Smart contract risk in the LRT contract. Each LRT protocol has its own contracts that manage deposits, AVS delegation, and reward distribution. A bug in those contracts is a direct loss vector. Most major LRTs have multiple audits, but audits do not eliminate risk. Smart contract exploits remain the single most common cause of catastrophic loss in DeFi.

Smart contract risk in EigenLayer. Underneath every LRT is the EigenLayer protocol itself. If EigenLayer has a bug that allows unauthorized withdrawal of restaked ETH, every LRT loses. EigenLayer has been audited extensively but, again, audits are not guarantees.

Oracle and depeg risk in DeFi. When LRTs are used as collateral on Aave or Spark, the lending protocol uses an oracle to value the LRT. If the oracle reads stale or manipulated prices, liquidations can cascade. The April 2024 ezETH event was driven by oracle latency: the secondary market price moved before the lending protocol oracle updated, triggering false liquidations.

Operator behavior risk. Even without explicit slashing, operators can underperform. Missed attestations reduce staking yield. Operator outages reduce AVS rewards. The LRT performs only as well as the operators it delegates to.

Regulatory risk. Restaking sits in a gray area for regulators. The classification of LRT yields and the legality of certain AVS reward structures could change. Several jurisdictions have begun looking at restaking the same way they look at staking-as-a-service, which has been challenged in some markets.

Smart contract risk in the leverage layer. If you use weETH on Aave with 5x leverage, you are exposed to the LRT contract, EigenLayer, Aave, the oracle, AND the dependencies of the wrapped variant. The risk does not add, it multiplies.

Frequently Asked Questions

Conclusion

Liquid restaking tokens are one of the most important DeFi primitives to emerge from the Ethereum staking ecosystem. They take the productivity of liquid staking, add the additional yield surface of EigenLayer restaking, and wrap everything into a single composable ERC-20 that plugs into Pendle, Aave, Curve, and the broader LRT-fi stack. For users who understand the layered risk profile, LRTs offer something genuinely new: a single asset that simultaneously secures Ethereum, secures third-party services, earns multiple reward streams, and remains usable across DeFi.

That said, LRTs are not a free lunch. The yield stack and the risk stack grow together. Slashing on AVSs becomes more real as the EigenLayer ecosystem matures. Operator concentration creates correlated risks that are easy to underestimate. Leverage on top of LRTs multiplies the downside as cleanly as the upside. And the post-EIGEN TGE era means headline APYs are coming down to more sustainable levels as the speculative points premium fades.

The right approach for most users is to treat LRTs the same way you would treat any layered DeFi exposure: understand each layer separately, diversify across protocols where possible, never lever up more than you can stomach during a depeg event, and keep an eye on the AVS landscape because that is where the next wave of risk and reward will come from. The LRT category is here to stay, but the winners will be the protocols that build durable AVS revenue, strong operator governance, and deep secondary liquidity. The participants who do best will be the ones who treat their LRT position as a portfolio decision rather than a yield chase. For a complementary perspective on how yield aggregators interact with this stack, our yield aggregator guide is a useful companion read.