What is Restaking? Complete 2026 Guide
Restaking is the practice of reusing already-staked ETH (or liquid staking tokens) to also secure other protocols — called "Actively Validated Services" or AVSs — and earn extra yield on top of base staking rewards. The category was created by EigenLayer in 2023 and has become a major piece of Ethereum's economic layer. In 2026, restaking yields have normalized from the early hype (typically 1-4% on top of ~3-4% base staking yield), the main risks (slashing across multiple AVSs, smart-contract risk in LRT protocols) are better understood, and the biggest LRT issuers are EtherFi, Renzo, Kelp, and Puffer.
This article is for educational and informational purposes only and does not constitute financial, investment, or trading advice. Cryptocurrency and DeFi investments carry significant risk, including the potential loss of all invested capital. Always conduct your own research (DYOR) and consult a qualified financial advisor before making any investment decisions. Past performance does not guarantee future results.
Key Insight
Restaking is the practice of reusing already-staked ETH (or liquid staking tokens) to also secure other protocols — called "Actively Validated Services" or AVSs — and earn extra yield on top of base staking rewards. The category was created by EigenLayer in 2023 and has become a major piece of Ethereum's economic layer. In 2026, restaking yields have normalized from the early hype (typically 1-4% on top of ~3-4% base staking yield), the main risks (slashing across multiple AVSs, smart-contract risk in LRT protocols) are better understood, and the biggest LRT issuers are EtherFi, Renzo, Kelp, and Puffer.
What is Restaking?
Restaking is the practice of taking ETH you already have at stake — securing Ethereum itself — and using it for double duty: also securing additional protocols and earning extra rewards from them.
The mechanism is simple. You stake ETH the usual way (32 ETH solo, or via an LST like Lido's stETH). Then you "restake" by opting in via a protocol like EigenLayer to also secure one or more Actively Validated Services (AVSs) — oracle networks, data availability layers, cross-chain bridges, app-specific rollups, and more. In return, you earn extra rewards from those AVSs on top of base ETH staking rewards.
The trade-off is risk. The original stake is exposed to slashing from Ethereum (if your validator misbehaves). With restaking, that same stake is now also exposed to slashing from each AVS you secure. More yield, more risk, same capital.
The Mental Model
Think of restaking as economic outsourcing of security. Every blockchain protocol — an oracle, a bridge, a rollup — needs some way to ensure its operators behave honestly. Traditionally that meant launching a token and bootstrapping a separate validator set. Restaking lets new protocols "rent" the security of Ethereum's already-staked ETH instead.
For the new protocol: instant access to deep security without launching a token. For the restaker: extra yield without locking up more capital. For Ethereum: a deeper economic moat around its base asset.
That value proposition created the category in 2023 and made it permanent by 2026.
The 2026 Restaking Stack
The stack has three layers:
1. Base staking
You stake ETH on Ethereum — solo (32 ETH minimum) or via a liquid staking token (Lido stETH, Rocket Pool rETH, Coinbase cbETH). This earns roughly 3-4% APR in 2026.
2. Restaking protocols
3. AVSs (Actively Validated Services)
The protocols that consume restaked security: oracles (e.g. Chainlink-style), data availability layers (e.g. EigenDA), cross-chain bridges, app-specific rollups, MEV networks, and more. By 2026 there are dozens of live AVSs producing rewards for restakers.
Liquid Restaking Tokens (LRTs)
For most retail participants, native restaking — running your own validator and opting into AVSs — is operationally too heavy. The dominant retail pattern is Liquid Restaking Tokens (LRTs).
An LRT issuer takes your ETH (or your LST), restakes it across a curated set of AVSs, and gives you back an LRT in return. The LRT accrues restaking yield and remains liquid — you can lend it, use it as collateral, or trade it like any other DeFi asset.
The major LRT issuers in 2026:
| LRT | Issuer | Notes |
|---|---|---|
| ----- | -------- | ------- |
| eETH | EtherFi | Largest LRT by TVL, native restaking |
| ezETH | Renzo | Strong DeFi integrations, recovered from a 2024 depeg episode |
| rsETH | Kelp | Multi-AVS exposure, strong tooling |
| pufETH | Puffer | Anti-slashing focus, novel operator design |
Choosing an LRT trades native-restaking yield (slightly higher) for liquidity (much more useful in practice). For most retail restakers in 2026, an LRT is the right answer.
For the broader stablecoin and tokenized-asset context, see our state of stablecoins 2026 and top RWA companies guides.
What Yields Actually Look Like in 2026
The early-2024 hype around 20%+ restaking yields did not materialize. As AVS supply caught up with demand, restaking rewards normalized:
| Component | Typical APR (June 2026) |
|---|---|
| ----------- | ------------------------ |
| Base ETH staking | 3.0-3.8% |
| EigenLayer restaking rewards | 0.8-2.5% |
| AVS-specific points/airdrops | Variable (often the largest historical component) |
| Total | ~4-8% APR |
Beyond the base figures, AVS-specific point programs and airdrops have historically been a meaningful (sometimes the largest) component of returns — though by 2026 these have normalized too, as most major AVSs have launched their tokens.
The Risks (read carefully)
Restaking has real risks that the yield numbers do not capture:
Slashing risk
If an AVS you secure misbehaves, you can lose a portion of your stake. With multiple AVSs, slashing exposure stacks. Operators that secure many AVSs simultaneously face correlated slashing risk if a common failure (a bug in a shared library, a coordinated attack) affects several at once.
Smart-contract risk
LRT protocols are complex. The Renzo ezETH depeg in 2024 was caused by a withdrawal-mechanism bug; while ezETH recovered, the episode was a real loss for holders during the dislocation. Bugs in LRT contracts, restaking protocols, or AVSs can all cause significant losses.
Peg risk
LRTs can de-peg from ETH in stress scenarios. The 2024 ezETH episode is the prominent example. In general, LRTs trade at a slight premium or discount to ETH depending on AVS yield expectations and liquidity dynamics. In a forced-withdrawal stress, those discounts can widen sharply.
Concentration risk
If a small number of operators secure most AVSs, correlated failures become more likely. The operator-set decentralization of the major restaking protocols has improved through 2025-2026 but remains a watch item.
Should You Restake?
The honest decision framework for 2026:
| Your situation | Recommended action |
|---|---|
| ---------------- | -------------------- |
| Already staking ETH, comfortable with DeFi risk, want a few extra % yield | Use an LRT (EtherFi, Renzo, Kelp, Puffer) |
| Already staking, want simplicity over yield | Stay in plain ETH staking or a major LST |
| Not yet staking, want lowest risk | Tokenized treasuries — see our BUIDL vs Ondo vs Hashnote |
| Running validators, willing to operate AVSs | Native restaking via EigenLayer |
| Want any yield without DeFi exposure | Yield-bearing stablecoins or tokenized treasuries, not restaking |
Native vs Liquid Restaking
Native restaking means you run a validator and opt into AVSs directly. You earn the full restaking yield with no LRT-protocol fee, but you take on the operational complexity and you lose liquidity (your ETH is locked).
Liquid restaking means using an LRT. The LRT issuer handles the operational complexity for a small fee (usually 5-10% of the restaking yield), and you get a liquid, composable token in return.
For 95%+ of retail participants in 2026, liquid restaking is the right choice. The yield gap is small; the convenience and composability are enormous.
Common Mistakes
- Treating LRT yield as risk-free. It is not. Slashing, smart-contract, and peg risks are real.
- Stacking many AVSs blindly. More AVSs means more correlated slashing exposure. The marginal yield is often not worth the marginal risk.
- Forgetting LST risk underneath. If you restake an LST, you carry the LST's risk plus the restaking risk. Pure-ETH LRTs avoid this.
- Ignoring withdrawal queues. Exiting a position is not instant. Plan for the LST and LRT withdrawal mechanics in advance.
Conclusion
Restaking turned into permanent Ethereum infrastructure between 2023 and 2026. The early hype faded, the yields normalized, the risks became better understood, and the category settled into a stable role: a way for ETH stakers to earn 1-4% extra yield in exchange for additional slashing and smart-contract exposure.
The 2026 playbook for retail:
- Already stake ETH? Consider adding an LRT for marginal extra yield
- Pick a major LRT issuer (EtherFi, Renzo, Kelp, Puffer) and understand its specific risk profile
- Size the position to your real risk tolerance — restaking is not free yield
- Track the AVSs your LRT exposes you to; concentration matters
For related Web3 yield and savings options, see our guides on staking ETH, tokenized treasuries, and stablecoins in 2026.
Key Takeaways
- Restaking is reusing staked ETH (or LSTs) to secure additional protocols — AVSs — and earn extra yield on top of base staking rewards
- EigenLayer created the category in 2023; by 2026 it remains the largest restaking protocol but is no longer the only one — Symbiotic, Karak, and others compete
- Total restaking yields in 2026 typically run 1-4% on top of ~3-4% base ETH staking yield — well below the early-2024 hype-driven projections
- Liquid Restaking Tokens (LRTs) — EtherFi eETH, Renzo ezETH, Kelp rsETH, Puffer pufETH — let you restake while keeping a liquid, composable token to use elsewhere in DeFi
- The main risks are real and stacked: slashing across multiple AVSs, smart-contract risk in LRT protocols, peg risk on the LRT itself, and concentration risk in operator sets
- In 2026 most retail restakers use an LRT (for liquidity) rather than native restaking (which requires running validators) — convenience usually beats the small yield premium
- Restaking is now mainstream-DeFi infrastructure — many production protocols rely on restaked security and the category is here to stay even as yields normalized
Frequently Asked Questions
What is restaking in simple terms?
Restaking is using your already-staked ETH for double duty. You stake ETH the normal way and earn staking rewards. Then you "restake" by opting in to secure additional protocols — AVSs (Actively Validated Services) — and earn extra rewards from those. The trade-off is more risk: if the AVS gets slashed for misbehavior, you can lose a portion of your stake. More yield, more risk, same capital.
What is EigenLayer?
EigenLayer is the original restaking protocol — the one that created the category in 2023. It runs a marketplace where ETH stakers can opt in to secure AVSs (oracle networks, data availability layers, bridges, app-specific rollups) and earn rewards. In 2026, EigenLayer is still the largest restaking protocol by TVL but no longer the only one. Symbiotic, Karak, and a small number of newer protocols compete in the same space.
What are LRTs (Liquid Restaking Tokens)?
LRTs are tokens that represent a deposit in a restaking protocol. You deposit ETH (or an LST like stETH), the LRT issuer restakes it for you across AVSs, and you receive an LRT in return — eETH from EtherFi, ezETH from Renzo, rsETH from Kelp, pufETH from Puffer. The LRT accrues restaking yield while remaining liquid and composable — you can lend it, use it as collateral, or trade it in DeFi. LRTs are how most retail restakers actually participate.
How much yield does restaking actually pay in 2026?
Total yields run 4-8% APR for most retail restakers in mid-2026: roughly 3-4% from base ETH staking plus 1-4% from restaking rewards on top. Yields vary by which AVSs you secure, how many you opt into, and how risk premiums move. The early-2024 hype around 20%+ restaking yields did not materialize as the AVS market scaled — supply caught up with demand and yields normalized.
What are the main risks of restaking?
Four big ones in 2026: (1) Slashing risk — if an AVS you secure gets slashed for misbehavior, you lose a portion of your stake. The more AVSs, the more correlated slashing exposure. (2) Smart-contract risk — LRT protocols are complex; bugs can cause catastrophic losses. (3) Peg risk — LRTs can de-peg from ETH in stress scenarios, as happened to ezETH briefly in 2024. (4) Concentration risk — if a small number of operators secure most AVSs, correlated failures become more likely. Restaking is not free yield.
Should I restake my ETH?
It depends on your risk tolerance and whether you already stake. If you already stake ETH and accept the extra slashing and smart-contract risk for an extra few percent yield, an LRT is a reasonable choice. If you want pure simplicity and lowest risk, plain staking (or an LST like Lido or Rocket Pool) is fine. If you do not want any DeFi exposure, a tokenized treasury product like our [BUIDL/Ondo/Hashnote comparison](/blog/tokenized-treasury-bills-blackrock-buidl-ondo-comparison-2026) covers may fit better. Match the product to your risk tolerance.
About the Author
Marcus Williams
Blockchain & DeFi Editorial Desk
Blockchain & DeFi Editorial Desk · Web3AIBlog
Marcus Williams is a pen name for our blockchain and DeFi editorial desk. Posts under this byline are written and reviewed by contributors with backgrounds in protocol engineering, on-chain analysis, smart contract auditing, tokenomics, and decentralized finance. The desk covers consensus mechanisms, liquidity protocols, MEV, on-chain forensics, regulatory frameworks across jurisdictions, and the operational realities of running and using DeFi at scale. We publish nothing about live protocols without testing on mainnet first.
