A liquid restaking token is a transferable claim on capital that has been restaked, so the holder is exposed to base staking rewards plus whatever the restaking layer pays, and to every slashing rule attached to both. It differs from a liquid staking token in what sits underneath. An LST is a claim on stake securing one chain; an LRT is a claim on stake securing that chain plus a set of services an operator selected. Historically part of what an LRT represented was points, which record participation rather than a contractual rate.
An LRT stacks a second protocol, a second operator relationship and an open-ended set of third-party slashing rules on top of an LST that already had its own. The yield is additive by design. The loss surface is additive by construction, and it is the part that never appears in the headline number.
Scroll to see the full diagram
The one structural difference between an LST and an LRT
Liquid Collective draws the line cleanly. A liquid staking token is a receipt token providing access to liquidity while participating in staking on a proof-of-stake network, programmatically minted to represent ownership of staked tokens and the network rewards they earned. A liquid restaking token is minted to provide liquidity while participating in restaking, and most LRTs represent ownership of restaked tokens, including liquid staking tokens deposited into EigenLayer to be restaked, and the EigenLayer points received.1
Chainlink describes the same object from the mechanics side: liquid restaking protocols pool user deposits, manage the delegation of those assets to node operators, and issue a tradable receipt representing the underlying deposits plus accumulated rewards.2 Every verb in that sentence names a decision the depositor is not making.
So the difference is not the yield and it is not the wrapper. It is the counterparty set. An LST holder's counterparties are a staking pool and its validators. An LRT holder's counterparties are that pool, plus a restaking protocol's contracts, plus an operator, plus every service that operator chose to secure. Four relationships behind one token balance.
Who is actually picking the risk
The academic survey is specific about the intermediary role: liquid restaking protocols act as intermediaries, managing validator operations and service selection on behalf of users, and issuing tokens that accrue value and can be traded or pushed into DeFi for further yield.3 Service selection is the phrase to sit with. The decision that determines which third-party slashing rules govern a depositor's capital is made by a protocol team, continuously, after the deposit.
None of that risk disappears when it is wrapped. The 2024 systematization describes restaking as multiplying slashing risks precisely because the stake is committed to several services at once.3 EigenLayer's own security documentation adds the protocol layer underneath: it names the possibility of a vulnerability that bypasses the withdrawal delay, the risk of a flaw in the code handling requests after the delay has run, and the fact that while contracts are paused the system permits arbitrary state or code changes through upgrades.5 An LRT removes none of that. It packages it and makes it transferable, which is genuinely useful and also makes the risk far easier to hold without having read it.
In our view the service-selection policy is the single most consequential document an LRT issuer publishes, and across the projects we have reviewed it is the one least likely to be versioned, governed or even written down. A yield page updated weekly and a risk policy updated never is a disclosure asymmetry, not an oversight.
Points, and what they actually are
Points sit inside the definition rather than beside it. Liquid Collective's description of what most LRTs represent includes the EigenLayer points received alongside the restaked tokens themselves.1 A holder buying an LRT during a points program is buying two things: a claim on restaked capital, and a running tally of participation.
That tally settles in a token. EigenLayer's governance token is EIGEN, and it is also used to pay restaking rewards to stakers.3 What a points program does not fix in advance is the conversion basis, the total to be distributed, the eligibility rules or the timing. Those are set later, by the issuer, and until they are set the point has no denominator.
Our position, labeled as opinion: a point is a claim on a future discretionary distribution being priced by a market before its terms exist. That is a legitimate thing to run and a legitimate thing to hold. It is not a yield, and folding an estimated points value into a quoted APY is a disclosure failure rather than an aggressive assumption. If you run a points program, publish the conversion basis before the secondary market prices one on your behalf.
The concentration nobody quotes
As of 3 August 2026, DefiLlama's public data API records the Liquid Restaking category at $4,345,818,119 across 30 tracked protocols. The distribution inside that total is ether.fi at $3,218,320,711, Kelp at $870,167,090, Renzo at $90,886,518, Puffer at $44,670,948 and Mantle Restaking at $31,765,659, with the remaining protocols below that.4
Do the division. One protocol holds about 74 percent of the tracked category and the top two hold about 94 percent. The methodology caveat has to travel with the number: these are DefiLlama's own protocol-level sums under DefiLlama's own category taxonomy, retrieved from the public data API because the web interface blocks automated clients, and both the classification and the figures move continuously.
The design consequence is a correction to a common risk statement. Spreading a position across three liquid restaking tokens sounds like diversification and mostly is not, because the second and third names are a rounding error against the first and because the underlying LSTs, operators and services overlap heavily. Concentration at the token layer is visible. Concentration at the operator and service layer, which is where slashing actually happens, is not published at all.
Scroll to see the full diagram
The stacked position, priced honestly
Take 100 ETH into an LRT and separate what you own from what you are exposed to. What you own is a receipt on pooled deposits that were staked, then restaked, then delegated. What you are exposed to, for a position securing three services, is Ethereum's consensus rules, the liquid staking pool's operator set and its loss-sharing rule, the restaking protocol's contracts and withdrawal delay, and three separate service-defined slashing conditions. One return line, six exposure lines.
The academic framing calls this the implicit leverage on restaked collateral: the stake is horizontally multiplied by securing additional services, and the slashing risks are multiplied with it.3 Leverage is the right word precisely because the position size did not change. Only the number of claims against it did.
The composite-rate arithmetic itself belongs to the restaking yield entry, which works through how a base rate, service payments and incentives combine. What belongs here is the structural point that the yield surface and the loss surface are the same surface counted twice, once with a plus sign on a marketing page and once with a minus sign in a risk document that most issuers have not written.
Exit runs through three queues, or one order book
The slow path out is sequential. The restaking protocol imposes its own withdrawal delay, which EigenLayer documents under the name escrow period and describes as a security measure for instances of vulnerability disclosure or when monitoring systems detect anomalous behavior.5 That is a circuit breaker, not a queue, and it is worth understanding as one: it is designed to stop capital leaving at exactly the moment a holder most wants it to leave. Underneath it, the liquid staking pool has a withdrawal queue of its own. Underneath that, Ethereum's validator exit queue lengthens with how many people are leaving at the same time. Three waits, stacked, each set by a different party.
The fast path out is the secondary market, priced by whoever is still bidding. That inherits the entire depeg mechanism documented on the liquid staking token entry, with an additional layer of uncertainty on top, because valuing an LRT means valuing service exposure that is not marked anywhere.
One concrete dependency shows how tightly the layers are coupled. Lido documents that its Ethereum mainnet Chainlink-compatible rate feed is deployed and used by Mellow LRT vaults, wrapping a call to wstETH.getStETHByWstETH, and that multiple Mellow LRT projects are built on top of stETH and wstETH.6 An LRT's valuation is running off the LST's rate feed. When that rate feed's assumptions are tested, the LRT is tested at the same instant, through a dependency most holders have never looked at.
What we hold LRT issuers to
Publish the service list with live status and the date it was last accurate. A list mixing live services with announced ones, undated, is the single most common defect we find in this category, and it turns a risk disclosure into a roadmap.
Quote the return as three separate lines rather than one number: the confirmed base staking rate, the payments from services that are actually live, and points, labeled as points, with their terms or an explicit statement that the terms do not exist yet. Then publish the three exit paths and their current lengths, and the loss-sharing rule that applies when a slashing event lands on one operator's delegators rather than the pool.
The anchor underneath all of it. An LRT is a distribution mechanism for security that services are renting, and if those services do not develop businesses that can pay for security out of revenue, the whole structure is being funded by incentives with a schedule. That is a question about the businesses, not about the token. This page is reference material for design work. It is not investment advice, not a recommendation about any protocol or asset, and it makes no claim about future returns.
Common questions
What is the difference between an LST and an LRT?
An LST is a receipt for staked tokens and their network rewards on one proof-of-stake network. An LRT is a receipt for restaked capital, and most LRTs represent ownership of restaked tokens including LSTs deposited into EigenLayer plus the points received.1 The practical difference is the counterparty set: an LRT adds a restaking protocol, an operator and every service that operator secures, each with its own slashing conditions, on top of everything the LST already carried.
How is LRT yield calculated?
It is a sum, not a rate. The base component is the staking reward on the underlying capital. On top sit payments from each service the operator secures, weighted by how much collateral is allocated to each, net of the operator's commission, and historically a points component with no fixed conversion basis. Because service participation changes over time, the composite moves. The full arithmetic belongs to the restaking yield entry rather than being repeated here.
What are restaking points?
Points are a record of participation, not a contractual return. Liquid Collective's description of what most LRTs represent explicitly includes EigenLayer points alongside the restaked tokens.1 EigenLayer's EIGEN token is used to pay restaking rewards to stakers.3 The conversion basis, total distribution, eligibility and timing are set by the issuer after the fact, so a point has no denominator while it is being earned. Treating an estimated points value as yield overstates a return that has no terms yet.
How big is the liquid restaking market?
DefiLlama's public data API records the Liquid Restaking category at $4,345,818,119 across 30 tracked protocols as of 3 August 2026, with ether.fi at $3.22 billion, Kelp at $870 million and Renzo at $91 million.4 That makes one protocol roughly 74 percent of the tracked total. These are DefiLlama's own category definitions and sums, they move continuously, and any figure quoted from them needs its date attached.
Can an LRT be redeemed for ETH immediately?
Not through the protocol. Redeeming means clearing the restaking protocol's withdrawal delay, which EigenLayer documents as an escrow period serving as a security measure for vulnerability disclosure or detected anomalous behavior, then the liquid staking pool's withdrawal queue, then Ethereum's validator exit queue.5 The last of those lengthens exactly when many people want out. The immediate route is selling on a secondary venue at whatever the market pays, which is a different transaction at a different price with no parity guarantee.
See LRT Tokenomics Guide for how this applies in practice.
Sources
- What is the difference between a liquid restaking token (LRT) and a liquid staking token (LST)?
Liquid Collective, 2024
Definitions of both token types and the statement that most LRTs represent ownership of restaked tokens including deposited LSTs and the EigenLayer points received. - Liquid Restaking: How LRTs Secure Web3
Chainlink, 2026
Mechanics of LRT issuance: pooling deposits, managing delegation to node operators, and minting a tradable receipt for deposits plus accumulated rewards. - SoK: Liquid Staking Tokens (LSTs) and Emerging Trends in Restaking
arXiv preprint 2404.00644v3, 2024
LRT protocols as intermediaries managing validator operations and service selection, LRTs as pegged claims rather than derivatives, the implicit-leverage and multiplied-slashing framing, and EIGEN as the token paying restaking rewards. - Liquid Restaking category TVL, public data API
DefiLlama, 2026
Category total of $4,345,818,119 across 30 tracked protocols read on 3 August 2026, with per-protocol sums for ether.fi, Kelp, Renzo, Puffer and Mantle Restaking. DefiLlama's own taxonomy; figures update continuously. - Withdrawal Delay
EigenCloud, Eigen Labs, 2026
The withdrawal delay, also called the escrow period, applied to LST tokens, the EIGEN token and native restaking as a security measure for vulnerability disclosure and detected anomalous behavior, plus the protocol's own caveats about bypass vulnerabilities, post-delay code flaws and arbitrary contract upgrades while paused. Cloudflare returns 403 to automated clients; verified live by stealth fetch on 3 August 2026. - Lido tokens integration guide
Lido Docs, 2026
Evidence of the LST to LRT dependency: Mellow LRT vaults use Lido's Chainlink-compatible mainnet rate feed wrapping wstETH.getStETHByWstETH, and multiple Mellow LRT projects are built on stETH and wstETH.
Last reviewed 2026-08
Know the terms but not sure how they apply to your project? That is what an engagement is for. We design, document, and stress-test the whole token economy inside the Tokenomics Data Room.
80+ projects advised. Complete tokenomics in 4 to 6 weeks.