Liquid restaking has emerged as the dominant pathway for ETH restaking exposure across 2024-2026, capturing approximately 77% of total restaked ETH. The major Liquid Restaking Tokens (LRTs) — EtherFi's eETH/weETH, Renzo's ezETH, Kelp DAO's rsETH, Puffer's pufETH, and several smaller alternatives — each operate with somewhat different economic structures, AVS exposure decisions, and realized yield outcomes. Q1 2026 provides 6 months of operational data with the framework substantially mature. The realized yield comparison across the major LRTs and native EigenLayer direct restaking is structurally informative for traders evaluating their own restaking positioning decisions.
I have been running positions across multiple LRTs and native EigenLayer through 2024-2026 and the realized Q1 2026 yield differentials provide specific insight into how the different pathways actually compare at scale.
The Q1 2026 Major LRT TVL And Yield Data
Major LRTs at the late-April 2026 observation window:
EtherFi (eETH/weETH):
- TVL: approximately $5.8 billion
- Q1 2026 average annualized yield: approximately 5.4-5.8%
- Yield decomposition: approximately 3.1% base ETH staking, approximately 2.3-2.7% restaking premium
- AVS exposure: diversified across approximately 8-12 active AVSs
- Operator selection: predominantly EtherFi-managed operators with some delegation
Renzo (ezETH):
- TVL: approximately $2.6 billion
- Q1 2026 average annualized yield: approximately 5.2-5.6%
- Yield decomposition: approximately 3.1% base ETH staking, approximately 2.1-2.5% restaking premium
- AVS exposure: diversified across approximately 6-10 active AVSs
- Operator selection: predominantly Renzo-coordinated operators
Kelp DAO (rsETH):
- TVL: approximately $1.4 billion
- Q1 2026 average annualized yield: approximately 5.4-5.8%
- Yield decomposition: approximately 3.1% base ETH staking, approximately 2.3-2.7% restaking premium
- AVS exposure: diversified across approximately 6-9 active AVSs
- Specific AVS rotation strategy with regular reallocation
Puffer (pufETH):
- TVL: approximately $1.1 billion
- Q1 2026 average annualized yield: approximately 5.0-5.4%
- Yield decomposition: approximately 3.1% base ETH staking, approximately 1.9-2.3% restaking premium
- AVS exposure: more conservative than other LRTs (approximately 4-7 active AVSs)
- Operator selection: includes specific Puffer-validated operators with anti-slashing measures
Native EigenLayer (direct deposit, depositor-managed AVS selection):
- TVL: approximately $0.9 billion
- Q1 2026 yield range: approximately 4.8-7.2% (broader range due to depositor-specific AVS choices)
- Mean depositor yield: approximately 5.2-5.6% (broadly comparable to LRT yields)
- Highly variable based on specific depositor AVS selection
The Realized Yield Differentials Decomposition
Three structural observations about the realized yield differentials.
First, the realized yield differentials between major LRTs are smaller than the marketing differentials would suggest. Each LRT markets specific yield-enhancement strategies, but the realized Q1 2026 yield range across the major LRTs is approximately 5.0-5.8% — a 0.8 percentage point spread. The realized differentials reflect different AVS exposure strategies and operator selection decisions, but the aggregate variation is bounded by the underlying restaking yield environment.
Second, native EigenLayer direct restaking produces broadly comparable mean yields to the LRTs but with materially higher variance. The realized native EigenLayer yield range of approximately 4.8-7.2% reflects depositor-specific AVS choices. Depositors who selected high-yield AVSs (typically with somewhat higher slashing risk) captured higher realized yield; depositors who selected conservative AVSs captured lower yield. The mean across the native EigenLayer depositor population is broadly comparable to the LRT means.
Third, Puffer's somewhat lower realized yield reflects its more conservative AVS exposure framework. The realized 5.0-5.4% yield on pufETH versus the 5.4-5.8% on eETH reflects approximately 0.4 percentage points of yield differential attributable to Puffer's more conservative AVS exposure decisions. Depositors who weight slashing protection heavily versus yield maximization may find this differential acceptable.
The Realized Slashing Track Record By LRT
Across the operational period through Q1 2026, the realized slashing events attributable to each major LRT:
- EtherFi: approximately $0.18 million cumulative slashing across all delegated operators (approximately 0.0031% of TVL)
- Renzo: approximately $0.06 million cumulative slashing (approximately 0.0023% of TVL)
- Kelp DAO: approximately $0.04 million cumulative slashing (approximately 0.0029% of TVL)
- Puffer: zero realized slashing events
- Native EigenLayer: approximately $0.12 million cumulative slashing across all native depositors (approximately 0.0133% of TVL)
The realized slashing rate on Puffer at zero events reflects its conservative AVS exposure framework. The native EigenLayer slashing rate of 0.0133% is materially higher than the major LRT rates because some native depositors selected aggressive AVS exposures that produced realized slashing.
For depositors weighting slashing risk heavily, Puffer's track record provides specific structural advantage at the cost of somewhat lower realized yield. For depositors with broader risk tolerance, the major LRTs provide somewhat higher realized yield at modest realized slashing rates.
The Token-Pricing-And-Yield Combined Performance
Beyond raw yield, the realized total return on LRT positioning depends on the relative pricing of the LRT tokens against ETH. The realized Q1 2026 LRT-versus-ETH pricing performance:
- eETH/weETH versus ETH: approximately +0.3% relative outperformance
- ezETH versus ETH: approximately -0.4% relative underperformance
- rsETH versus ETH: approximately +0.1% relative neutral
- pufETH versus ETH: approximately +0.2% relative outperformance
The realized pricing differentials are small but cumulative across multi-quarter horizons. For depositors evaluating LRT selection, the realized total return profile combines yield (approximately 5.0-5.8% annualized) with pricing performance (approximately ±0.5% annualized differential against ETH).
The pricing differentials reflect specific LRT-pool dynamics including liquidity provision economics, redemption queue dynamics, and broader DeFi integration patterns. EtherFi's modest outperformance reflects its broader DeFi integration depth; Renzo's modest underperformance reflects specific liquidity pool dynamics during the Q1 vol regime.
What This Tells Me About Restaking Pathway Selection
Three structural reads from the realized Q1 2026 data.
First, for most depositors, the LRT selection is approximately yield-neutral across the major options. The realized yield differential between EtherFi, Renzo, Kelp DAO at approximately 5.4-5.8% is small enough that selection should be driven by other factors (DeFi integration depth, governance preferences, redemption pathway preferences) rather than by yield optimization specifically.
Second, Puffer provides specific structural advantage for slashing-conscious depositors. The combination of conservative AVS exposure, dedicated anti-slashing operator framework, and zero realized slashing events through the operational period produces specific structural value for depositors who weight slashing risk above yield maximization. The trade-off is approximately 0.4 percentage points of yield.
Third, native EigenLayer direct restaking is appropriate for sophisticated depositors with strong AVS selection conviction. The realized 4.8-7.2% yield range reflects substantial depositor-specific outcomes — sophisticated depositors capturing the upper end can outperform LRT yields meaningfully, while less-informed depositors may underperform LRT yields. Native restaking is operationally meaningful for depositors with strong analytical conviction about specific AVS selection.
My Current LRT Allocation
For my own restaking exposure, I run approximately the following allocation:
- EtherFi (eETH/weETH): approximately 40% of my restaking exposure
- Native EigenLayer with deliberate AVS selection: approximately 30%
- Kelp DAO (rsETH): approximately 15%
- Puffer (pufETH): approximately 10%
- Renzo (ezETH): approximately 5%
The mix reflects my specific operational considerations rather than yield maximization. EtherFi for the broadest DeFi integration; native EigenLayer for sophisticated AVS exposure with my own analytical conviction; Kelp DAO and Puffer for diversification.
For traders evaluating their own restaking allocation, the realized data supports diversified allocation across major LRTs rather than concentration in any single LRT. The realized yield differentials are small enough that diversification benefits exceed yield optimization benefits for most depositors.
Honest Limits
I did not access protocol-specific operational data — the TVL, yield, and slashing figures referenced here come from publicly disclosed protocol reports and DeFi Llama aggregations through April 2026. The yield decomposition between base ETH staking and restaking premium reflects approximate calculations from publicly disclosed yield data. The slashing event attribution to specific LRTs reflects publicly known events and may not capture undisclosed operational issues. The personal allocation observations reflect my own current restaking exposure and are not investment advice or recommended allocation. Individual depositor risk tolerance and operational understanding affect appropriate restaking pathway selection. The realized restaking ecosystem may evolve through 2026-2027 as additional AVSs onboard or as protocol frameworks change materially.