Best Restaking Platforms 2026, Ranked by TVL

Last updated: 23 October 2026. TVL refreshed from DefiLlama, the April 2026 rsETH bridge exploit added to the Kelp entry, and ether.fi's exit timeline updated.

By size, the leading restaking platforms in October 2026 are EigenCloud (formerly EigenLayer, about $6.2B), Babylon for bitcoin (about $3.3B) and Kelp's rsETH for liquid restaking (about $0.94B), per DefiLlama on 23 October 2026. Symbiotic now presents itself as a collateral market, Karak has become OpenGDP, and ether.fi is leaving EigenLayer restaking. The "best" restaking platform is the one whose slashing conditions you can name and whose fees, not points, justify them; for many holders that means plain liquid staking instead.

Restaking took an idea everyone understood (stake tokens, secure a network, earn rewards) and added a second layer: use that same staked position to secure additional services, and earn again. More yield from the same capital.

The mechanism is real. So is the risk, and it is not the same risk as staking. And in 2026 the market has changed: slashing is live, the biggest names have rebranded, and much of the capital has left. This guide compares the platforms that remain, then explains how to judge any of them.

Educational only, not financial advice.

Restaking platforms compared (October 2026)

Figures are from DefiLlama on 23 October 2026. Liquid restaking tokens mostly deposit into EigenCloud, so the numbers overlap and should not be added together.

Best restaking platform by use case

• Best for ETH holders who want to pick operators directly: EigenCloud.
• Best for bitcoin holders: Babylon, the only venue at scale that stakes BTC without bridging it.
• Best for a single liquid token: Kelp's rsETH is still the largest liquid restaking token, but read the note on its April 2026 bridge exploit below.
• Best if you mainly want staking yield: a plain liquid staking token, which skips the extra slashing layer; see the alternative below.

The pattern is the story. Every major venue is well below its peak, and the decline is not only price: capital has left because the fees restaking pays have not justified the risk.

The platforms, one by one

EigenCloud (formerly EigenLayer)

Still the largest restaking venue by a wide margin. EigenLayer now operates as EigenCloud and markets verifiable computing, with restaked collateral as the security layer underneath (Blockhead, 29 September 2026). Slashing went live on mainnet on 17 April 2025 and is opt-in: services must adopt it, and operators can attribute stake to specific services so that one faulty service cannot reach all of it (CoinDesk, 17 April 2025). Best for: ETH holders or solo stakers who want to choose operators directly rather than through a wrapper.

Babylon

The bitcoin option. Babylon lets BTC holders stake to secure other networks without bridging their bitcoin to another chain (DefiLlama). TVL has roughly halved from its October 2025 peak but it remains the second-largest venue. Best for: bitcoin holders who want staking-style exposure and accept a newer, less battle-tested security model.

Kelp (rsETH)

The largest liquid restaking token. Kelp restakes on your behalf and gives you rsETH, which is usable across DeFi. Convenient, but you hold staking, restaking, contract and market-price risk in one token, and Kelp chooses the operators. That wrapper risk became real on 18 April 2026, when an attacker drained about 116,500 rsETH (roughly $292M) through Kelp's LayerZero-based cross-chain bridge, which relied on a single verifier, and used unbacked rsETH as collateral on Aave before Kelp paused its contracts and Aave froze rsETH markets (The Block). Nothing was slashed; the loss came from the bridge around the token. Best for: users who want restaking exposure as a single liquid token and will check its secondary market depth.

Symbiotic

Launched as a restaking rival to EigenLayer; DefiLlama now classifies it as a collateral markets protocol, describing capital "committed to financial obligations and enforced by code" (DefiLlama). ether.fi users who still want restaking exposure must opt into a separate product built on Symbiotic (Blockhead).

Renzo (ezETH)

Once one of the biggest liquid restaking tokens, now about 3% of its peak size. Renzo is also the clearest case study in liquid restaking risk: on 24 April 2024 ezETH traded as low as about $700 on some markets, a 79% drop in under an hour, and more than 250 users lost around $56M to liquidations (DL News, 24 April 2024). Nothing was slashed; the loss came from the wrapper and the leverage on top of it.

OpenGDP (formerly Karak)

Karak was one of the largest restaking layers in mid-2024. Under its new name, its tracked TVL was about $7M on 23 October 2026, against a peak above $1B (DefiLlama).

What about ether.fi?

ether.fi was historically one of the largest restaking depositors. In August 2026 it removed restaking from weETH, moving restaking exposure into a separate Symbiotic-based token, and in late September it said it would complete its exit from EigenLayer by the fourth quarter, with less than 1% of assets still restaked as of August (The Defiant); CEO Mike Silagadze said "there were no meaningful yield opportunities in restaking" (Blockhead, 29 September 2026). Its eETH is now essentially a liquid staking token, covered in our best crypto staking platforms comparison.

The 2026 problem: restaking fees collapsed

The case for restaking is that services pay for the security your stake provides. In practice, that payment has been small. In a week in early September 2026, about $10.02B of restaked capital generated roughly $99,977 in fees, while about $51.87B in liquid staking generated about $27.35M, so ordinary staking earned around 53 times more per dollar (Blockhead, 29 September 2026). Much of the headline restaking yield of 2024 was points and airdrop expectations, not service fees.

That is the single most important input into choosing a platform: if the real premium is near zero, the additional slashing risk is being taken for free.

How restaking works

In ordinary proof-of-stake you bond tokens to a validator. Behave, earn rewards; misbehave, get slashed.

Restaking lets that bonded stake be pledged a second time as economic security for other services: oracles, bridges, data availability layers, sequencers. Those services pay for the security. You collect both the base staking reward and the service fees.

Liquid restaking adds a token on top: you deposit into a protocol, receive a transferable receipt representing the restaked position, and can use that receipt elsewhere in DeFi while it keeps earning.

The risk nobody prices properly

Every service your stake secures adds a slashing condition. Restake across five services and you are exposed to five independent sets of conditions, each written by a different team, each with its own bugs. Since April 2025 these conditions are enforceable on EigenCloud, not theoretical.

Three specific problems:

Correlated slashing

If several services share infrastructure (the same oracle feed, the same client software) one failure can trigger slashing across all of them simultaneously. The risks are sold as diversified and are frequently correlated.

Yield that does not compensate

An extra 2% for taking on an unaudited service's slashing conditions is not obviously a good trade. Most restaking yields are paid in the service's own token, whose value is correlated with the thing you are underwriting.

Unbonding under stress

Exit queues lengthen exactly when everyone wants out. A liquid restaking token trading below par during stress is the market telling you the exit is not free.

How to read a restaking yield quote

Restaking yields are quoted as a single headline number that bundles several components with very different reliability. Decomposing it is the difference between comparing platforms and comparing marketing.

• Base staking reward. Protocol issuance for securing the underlying chain. The most predictable component, and the one you would earn without restaking at all.
• Service fees. Payments from the services your stake is securing. This is the actual restaking premium, and it depends on real demand for that security, which in 2026 has been far lower than projections assumed.
• Token incentives. Emissions paid in a platform or service token. Denominated in something whose price is unlikely to be stable, and the component most likely to fall sharply once an incentive programme matures.
• Points and expected airdrops. Not yield. A speculative claim on a future distribution with no defined value, frequently presented adjacent to real APR in a way that implies equivalence. See real yield for the distinction.

Ask what the yield is with the last two components at zero. If the answer is close to the base staking rate, you are being paid the ordinary rate to take an additional layer of slashing risk. That is a poor trade regardless of the headline.

Exit liquidity is the risk that actually bites

Restaked positions typically carry the underlying chain's unbonding period plus the additional withdrawal delay of each service securing the stake. Those delays compound, and they are longest exactly when you most want out.

Liquid restaking tokens are marketed as the answer, and they do solve the problem while secondary market depth holds. The mechanism to understand is that the token's peg is a market price, not a redemption guarantee: during stress, holders who want immediate exit sell into a thin book and the token trades below the value of the underlying position. You can exit at a discount, or wait out the queue. What you cannot do is redeem instantly at par. The ezETH episode above is what that looks like with leverage on top; our explainer on DeFi loan liquidations shows why loops unwind so fast.

Before depositing, check three things: the total unbonding path end to end, the actual secondary depth for the liquid token rather than its notional market cap, and whether the discount has widened during past volatility.

Sizing a position you can defend

Correlated slashing is the risk restaking introduces and ordinary staking does not: one operator fault can penalise your stake across every service it secures simultaneously. Position sizing should assume that correlation rather than treating each service as independent.

A conservative approach treats the whole restaked allocation as a single exposure to its worst-case slashing event, sizes it accordingly, and diversifies operator rather than service. Concentrating with one operator across many services maximises the correlated case: the arrangement that looks most diversified on a dashboard while being least diversified in reality.

What to check before depositing

1. How many services is your stake securing? More is not better.
2. What are the actual slashing conditions? If they are not documented plainly, that is the answer.
3. Who decides which services get your stake? Delegated curation means someone else is choosing your risk.
4. What is the unbonding period, and what happened last time there was a rush?
5. Is the yield paid in something you want to hold?
6. Has the contract been audited, and is the audit recent relative to the last upgrade?
7. Is the platform still growing or shrinking? Compare current TVL to its peak; a venue that has lost most of its capital may cut support for services or tokens you rely on.

Our guide to restaking explained covers the mechanics in more depth, and crypto risk management rules sets out how to size positions like this.

The simpler alternative: liquid staking

Restaking is one answer to "my staked capital is idle". Liquid staking is another, and it carries a single set of slashing conditions rather than several. Our guide to whether liquid staking is safe covers the risks it still carries.

You stake, receive a liquid token representing the position, and that token remains usable across DeFi while the underlying keeps earning validator rewards. No additional services, no stacked slashing.

JewelSwap runs this model across three networks with a dual-token design: mint the base liquid staking token 1:1 with your deposit (the design is meant to maintain 1:1 backing, and any extra minted through Protocol-Owned Liquidity goes to the protocol's own liquidity, not the user), then stake that for an appreciating variant:

• Sui: JWLSUI and SJWLSUI, covered in dual-token liquid staking on Sui
• MultiversX: JWLEGLD and SJWLEGLD, see how to stake EGLD
• Radix: JWLXRD and SJWLXRD, see JWLXRD liquid staking on Radix

The mechanism, including the 10-day unbonding and transferable claim NFT, is set out in what is liquid staking and derivative tokens explained. JewelSwap does not offer restaking: your stake secures its own chain and takes on no third-party slashing conditions.

Choosing between them

Restaking suits people who understand slashing conditions well enough to price them, and who are being paid enough to take that risk. That is a narrower group than the marketing suggests, and in 2026 it is narrower still.

For most holders the sensible comparison is between liquid staking options rather than between liquid staking and restaking. We compare the main options across chains in best crypto staking platforms of 2026 and the Sui landscape in Sui liquid staking compared, and if you want the yield without the extra slashing surface, how to stake SUI is the place to start.

The general principle from our risk rules applies: if you cannot explain in one sentence what would cause you to lose principal, you are not ready to size the position.

Look up unfamiliar terms in the JewelSwap crypto glossary.

Frequently asked questions

What is the best restaking platform in 2026?

By size, EigenCloud (formerly EigenLayer) leads at about $6.2B, followed by Babylon for bitcoin at about $3.3B and Kelp's rsETH at about $0.94B, per DefiLlama on 23 October 2026. The best choice is the one whose slashing conditions you understand and whose fees, rather than points, pay for the extra risk.

Is EigenLayer still the biggest restaking platform?

Yes. Now branded EigenCloud, it held about $6.2B on DefiLlama on 23 October 2026, more than any other restaking venue, though well below its August 2025 peak of about $22.1B. Slashing has been live on mainnet since 17 April 2025 and is opt-in per operator set.

Is ether.fi still a restaking platform?

Not for much longer. ether.fi removed restaking from weETH in August 2026 and in September said it would complete its exit from EigenLayer restaking by the fourth quarter, saying there were no meaningful yield opportunities in restaking. Users who want restaking exposure must opt into a separate Symbiotic-based product.

Was Kelp's rsETH hacked?

Kelp's rsETH bridge was exploited on 18 April 2026: an attacker took about 116,500 rsETH, roughly $292M, through its LayerZero-based cross-chain bridge and borrowed against it on Aave before contracts were paused. The loss came from the bridge setup, not from restaking slashing.

What is restaking in simple terms?

Using an already-staked position to secure additional services beyond the base chain, earning a second stream of rewards from the same capital. The additional services impose their own conditions, which means additional ways your stake can be penalised.

Is restaking riskier than staking?

Yes, and the difference is structural rather than a matter of degree. Ordinary staking exposes you to one slashing condition on one chain. Restaking exposes the same capital to every service's conditions at once, and a single operator fault can trigger several simultaneously.

What is correlated slashing?

When one event (an operator misconfiguration, a client bug, an infrastructure outage) breaches the conditions of multiple services your stake secures at the same time. Losses that look independent when modelled separately arrive together, which is why per-service risk assessment understates the true exposure.

Are liquid restaking tokens as liquid as they claim?

While markets are calm, largely yes. Under stress the peg is a market price supported by secondary depth, not a redemption right, so tokens can trade below the value of the underlying position precisely when holders want to exit. Renzo's ezETH fell about 79% in under an hour in April 2024. Check real book depth rather than market capitalisation.

Should I restake or just stake?

Decompose the yield first. If the restaking premium net of token incentives and speculative points is small relative to the base staking rate, you are accepting a materially larger risk set for a marginal return. In September 2026 liquid staking earned about 53 times more in fees per dollar than restaking. Many holders are better served by liquid staking, which keeps liquidity and adds no extra slashing surface.

How long does it take to withdraw?

The chain's unbonding period plus each service's withdrawal delay, which compound. JewelSwap's liquid staking uses a ten-day unbonding period with a transferable claim NFT, so the position can change hands while unbonding rather than being locked to one address.