The most common question I get from people trying to optimize ETH yield positioning is: which LRT should I hold? Multiple Liquid Restaking Tokens compete for the same user — ETH holders willing to accept restaking exposure for yield premium over plain LST. The four main contenders are ether.fi (weETH), Renzo (ezETH), Puffer (pufETH), and Kelp (rsETH). Each has different yield profile, different risk distribution, different DeFi composability depth.
This guide walks through what each does, where each wins, and gives my honest current recommendations for different user profiles.
The Sector Snapshot
Q1 2026 LRT sector total TVL is approximately $14-15B. The distribution is heavily concentrated:
ether.fi (weETH): approximately $11B TVL — sector dominant Renzo (ezETH): approximately $1.6B TVL — established second Kelp DAO (rsETH): approximately $0.7B TVL Puffer Finance (pufETH): approximately $0.4B TVL Smaller LRTs (Eigenpie, Swell rswETH integrations, others): combined $0.8-1.2B
ether.fi captured roughly 75% of the LRT sector during 2024 when restaking narrative was hottest and held that lead through 2025-2026 despite competitor attempts. Renzo, Puffer, Kelp each occupy specific positioning niches but operate at materially smaller scale.
ether.fi (weETH) — The Default Choice
ether.fi is essentially the LRT equivalent of Lido in LST. Largest TVL, deepest DeFi integration, longest operational track record (since 2023 in current form), strong distribution through CEX integrations (Coinbase Wallet, Binance Web3 Wallet support).
Q1 2026 weETH metrics: - Total APY: 3.5-4.5% (base ETH staking + EigenLayer AVS rewards + ether.fi specific incentives) - Native staking yield: ~3.0-3.2% from ETH staking - AVS rewards from EigenLayer: ~0.3-0.7% - ETHFI/KING token rewards: ~0.2-0.6% - DeFi integration depth: deepest among LRTs (Aave V3, Pendle, Morpho, Curve all support weETH) - Operational track record: ~30 months without major incidents
The DeFi composability is the biggest practical differentiator. weETH is accepted as collateral on Aave V3 across multiple chains, has Pendle PT/YT markets with substantial liquidity, integrates with Morpho vaults, supports Curve liquidity provision. For users wanting LRT positioning that composes with broader DeFi, weETH is structurally easier.
Pick weETH if: you want LRT exposure with maximum DeFi composability, you prefer largest LRT for liquidity depth, you value longest operational track record. This is the default choice for most users.
Renzo (ezETH) — The Multi-Chain Specialist
Renzo took different positioning by emphasizing multi-chain availability and native deployments on multiple L2s. ezETH is available on Ethereum mainnet, Arbitrum, Base, Linea, Mode, BNB Chain.
Q1 2026 ezETH metrics: - Total APY: 3.5-4.3% (similar magnitude to weETH) - Native staking yield: ~3.0-3.2% - AVS rewards: ~0.3-0.6% - REZ token rewards: variable - Multi-chain availability: broadest among major LRTs - Operational history includes April 2024 depeg event (briefly traded at 95% of ETH peg before recovery)
Renzo's April 2024 ezETH depeg event remains historically relevant. ezETH briefly traded at 5-10% discount to ETH peg following protocol decisions about restaking rewards distribution. The depeg recovered within days but caused liquidation cascades on protocols holding ezETH as collateral. Some users haven't fully forgiven Renzo for the operational stress.
Pick Renzo if: you specifically want LRT positioning native on Arbitrum, Base, or other supported L2s without bridging weETH from Ethereum. The multi-chain advantage matters for users with L2-concentrated DeFi positioning.
Puffer Finance (pufETH) — The Anti-Slashing Architecture
Puffer differentiated through technical architecture focused on slashing risk reduction. The Puffer protocol uses specific cryptographic techniques (secure-signer infrastructure) to limit slashing exposure for restaked ETH compared to standard restaking deployment.
Q1 2026 pufETH metrics: - Total APY: 3.4-4.2% - Slashing exposure structurally lower than alternatives (claim by Puffer; not fully validated by realized slashing events since none have occurred at scale) - DeFi composability: meaningful but smaller than weETH - Operational track record: shorter than weETH and Renzo
Puffer's structural pitch is genuine but the practical advantage hasn't been tested. Major slashing events on EigenLayer haven't occurred. Whether Puffer's architecture would actually outperform alternatives during slashing event remains unproven empirically.
Pick Puffer if: you specifically prioritize minimum slashing exposure as design principle and accept slightly smaller ecosystem in exchange for the architectural commitment. Smaller user base but legitimate technical positioning.
Kelp DAO (rsETH) — The Pioneer Multi-Asset LRT
Kelp emphasized supporting multiple LST types as backing rather than only standard ETH or stETH. rsETH backing includes wstETH, sfrxETH, ETHx (Stader), and other LST variants. The diversified backing was Kelp's structural pitch.
Q1 2026 rsETH metrics: - Total APY: 3.4-4.3% - Multi-LST backing structure - DeFi composability: moderate, between weETH and Puffer - KEP token rewards plus Kelp Miles incentive program - Operational history includes 2024 minor governance issues that resolved
Kelp's multi-LST backing is intellectually interesting but practical advantage is bounded. Most users supplying ETH to LRT don't care which specific LST backs the position — they care about yield, peg stability, and composability. The diversified backing arguably reduces concentration risk on Lido stETH but the actual concentration risk hasn't manifested as material problem.
Pick Kelp if: you specifically value the diversified LST backing or want to support smaller LRT for ecosystem diversification reasons. Smaller scale than alternatives but functional product.
The Yield Comparison
Honestly, the realized yield differences between these four LRTs are small. All deliver 3.4-4.5% APY range. Specific monthly performance varies by 10-30 basis points typically. No LRT systematically delivers materially higher yield over multi-month periods.
If you're choosing between LRTs based purely on yield, you're optimizing for noise. The 30bps differential isn't meaningful enough to drive selection. Other factors (DeFi composability, operational track record, specific feature preferences) should drive choice.
The Realized Risk Distribution
LRTs share common risks: ETH staking risk (slashing), EigenLayer AVS slashing risk, smart contract risk on LRT layer, peg risk relative to ETH, governance risk on LRT protocol.
Realized 2024-2026 incident history: - ether.fi: no major incidents - Renzo: April 2024 ezETH depeg event (brief, recovered) - Puffer: no major incidents - Kelp: minor 2024 governance issues, resolved
Based on realized history, ether.fi and Puffer have cleaner records. Renzo's depeg event remains meaningful for risk-averse users. Kelp's smaller governance issues were minor but worth noting.
My Personal Allocation
For my own LRT positioning across ~10-15% of ETH allocation:
ether.fi weETH: ~70-80% of LRT allocation. Default position based on operational track record and DeFi composability.
Renzo ezETH: ~10-15%. Used specifically for L2-native positioning on Arbitrum and Base where ezETH availability matters.
Puffer pufETH: ~5-10%. Smaller position for ecosystem diversification.
Kelp rsETH: minimal. Tested but didn't size meaningfully.
The concentration on weETH reflects the structural advantages plus my preference for established positions over diversification-for-diversification's-sake.
Recommended Picks By User Profile
For users wanting maximum LRT yield with minimum operational complexity: ether.fi weETH only. Single LRT, deepest ecosystem, easiest UX.
For users wanting LRT positioning across multiple L2s: Renzo ezETH for L2-native availability, weETH on Ethereum mainnet.
For users prioritizing slashing risk reduction: Puffer pufETH for the architectural commitment, sized small if you want established history.
For users wanting LRT diversification across multiple protocols: 70% weETH, 15% Renzo, 10% Puffer, 5% Kelp. Captures sector exposure with ether.fi concentration.
For users with stronger conviction in specific LRT thesis: concentrated positioning in your preferred LRT, sized within your overall risk budget.
For users new to restaking: start with weETH only. Get familiar with LRT mechanics before adding other LRTs to position.
A Note On Yield Premium Reality
The honest assessment is that LRT premium over plain LST has been smaller than the 2024-era restaking thesis projected. weETH at 3.5-4.5% APY captures roughly 50-150bps premium over Lido stETH at ~3.0-3.2%. The original thesis suggested 200-500bps premium would emerge as EigenLayer AVS demand matured.
The realized 50-150bps premium is real but bounded. For users evaluating whether LRT positioning makes sense vs plain LST, the additional yield is meaningful but not transformative. The trade-off includes additional smart contract risk (LRT layer plus EigenLayer base), peg risk, governance risk on LRT protocols.
I run substantial LRT exposure but appropriate sizing reflects the modest premium. Larger LRT concentration would require conviction that AVS economics improve materially, which is possible but not yet demonstrated.
Sourcing notes: TVL, yield, and operational figures from ether.fi, Renzo, Puffer, Kelp dashboards plus DefiLlama and EigenLayer ecosystem analytics through April 2026. TVL fluctuates ±15% across the quarter. Yield calculations approximate daily-rate annualized. The competitive comparison uses publicly available metrics that may use different methodology. Personal positioning observations reflect my own approach and aren't recommended allocations. LRT smart contract risk, EigenLayer slashing risk, peg risk, and governance risk all apply. None of this is financial advice.