Staking ETH: Solo, Liquid, CEX, and DVT

State of ETH staking in 2026

~39.1M ETH (32% of supply) is staked across ~898K active validators in May 2026. Base consensus yield is ~2.73–2.84%; with MEV and priority fees, solo validators earn roughly 3.0–3.3% total. Last verified: 2026-05-27.

Post-Pectra (Pectra is the Ethereum hard fork combining Prague execution-layer and Electra consensus-layer changes, activated May 7, 2025, raising the validator max effective balance from 32 to 2,048 ETH via EIP-7251), which raised the max effective balance to 2,048 ETH, the question isn't whether to stake ETH — it's which of the four standard methods fits your stack. Yields are compressed from the early 5%+ post-Merge windows (more validators competing for the same issuance), but the tooling has matured. Lido, Rocket Pool, and Ether.fi cover 99% of retail use cases; solo staking via DappNode is appliance-grade; and the restaking layer (EigenLayer, Symbiotic, Karak) added a second income stream at the cost of an additional slashing surface.

~39.1M ETH is staked — roughly 32% of total supply — secured by ~898K active validators (validatorqueue.com, beaconcha.in, May 2026). Base consensus-layer yield is ~2.73–2.84%; with MEV and priority fees from the execution layer, total solo validator returns run ~3.0–3.3%. Staking penetration has grown rapidly into 2026, driven by spot ETH ETF staking distributions now live at Grayscale and expected from BlackRock and Fidelity. More stake = lower per-validator yield — exactly the dynamic Ethereum's issuance curve is designed for, and the reason yields will continue compressing as staking penetration approaches 35–40%.

The validator entry queue is under significant pressure in May 2026: approximately 3.3M ETH is queued with a wait time of roughly 57 days (churn limit: 256 ETH per epoch / ~57,600 ETH per day), driven by institutional and ETF-related staking demand. The exit queue is near zero (~4-day wait). New solo validators face a long activation delay; liquid staking deposits are unaffected.

Three rules: (1) under 32 ETH, use Lido or Rocket Pool; (2) solo staking pays ~0.3–0.6% more gross than liquid staking — do it for decentralization, not yield; (3) CEX staking surrenders 25–30% of yield to the exchange on top of counterparty risk — only worthwhile if the ETH was already sitting there.

ETH staking explained in 30 seconds

Staking ETH locks your coins as validator collateral; the validator earns ETH for proposing and attesting blocks. Four methods exist: solo (32 ETH), liquid (any amount), CEX (one click), and DVT (split validator). Last verified: 2026-05-27.

Staking (Ethereum staking is the act of locking ETH as collateral for a validator that secures the proof-of-stake network) locks ETH as collateral for an Ethereum validator. The validator participates in consensus (proposing blocks when selected, attesting every epoch ~6.4 minutes) and earns rewards from three sources: consensus-layer issuance (the predictable ~2.73–2.84% base), execution-layer priority fees (variable, tx-volume dependent), and MEV (Maximal Extractable Value is the profit validators capture by reordering or including transactions, typically extracted via MEV-Boost relays) (variable, captured via MEV-Boost). Staked ETH is at risk of "slashing" (slashing is the protocol penalty that destroys part of a validator's stake for misbehavior such as double-signing or surround-voting attestations) if the validator misbehaves — but slashing only triggers on specific, intentional-looking faults (double-signing, surround votes); normal downtime costs only missed attestation rewards.

There are four main ways to stake:

1. Solo staking (32 ETH required, max decentralization, ~3.0–3.3% with MEV)
2. Liquid staking (any amount — Lido stETH ~2.4%, Rocket Pool rETH ~2.1%, Ether.fi eETH ~2.4%)
3. CEX staking (Coinbase, Kraken, Binance — simplest, lowest net yield, counterparty risk)
4. DVT staking (SSV, Obol — splits a validator across multiple operators for fault tolerance)

The four staking methods

Solo staking pays ~3.0–3.3% gross with 32 ETH; liquid staking pays 2.1–2.4% net from 0.01 ETH; Rocket Pool megapool validators need 4 ETH bonded; CEX staking yields ~2.1–2.6% net after the exchange's 25–30% cut. Last verified: 2026-05-27.

Method 1: Solo staking (most decentralized)

Solo staking (running your own Ethereum validator with 32 ETH and no third-party intermediary) requires 32 ETH and a 24/7 machine. It earns ~3.0–3.3% gross (no protocol fees, full MEV) but you carry all hardware and operational risk. Last verified: 2026-05-27.

Requirements:

• 32 ETH minimum (post-Pectra, you can consolidate validators up to a single 2,048 ETH balance and compound rewards without spinning up new validators)
• A computer running 24/7 — Raspberry Pi 5 with NVMe SSD ($300), Intel NUC ($800), or any modern x86 box. 16–32 GB RAM, 2TB+ NVMe SSD is the realistic floor in 2026.
• Stable internet — 25Mbps down / 5Mbps up sustained is sufficient
• Basic Linux familiarity (editing YAML, restarting systemd services, monitoring logs)

Setup (roughly):

1. Set up an execution client — Geth, Nethermind, Reth, Erigon, or Besu. Nethermind and Reth are the recommended minority clients in 2026; pair with a minority consensus client to support client diversity.
2. Set up a consensus client — Lighthouse, Prysm, Teku, Nimbus, or Lodestar.
3. Generate validator keys with staking-deposit-cli on an air-gapped machine for any serious amount.
4. Deposit 32 ETH via the official Ethereum launchpad at launchpad.ethereum.org — verify the URL before signing.
5. Run a validator client signing your keys. Use Web3Signer with a remote keystore for key isolation.
6. Wait for activation — currently ~57 days in the entry queue (May 2026 surge; this will vary). Run MEV-Boost with at least two relays (Flashbots, BloXroute, Aestus, Ultra Sound) from day one.

Tools that simplify it:

• DappNode — turnkey appliance OS with dropdown setup instead of CLI
• eth-docker — docker-compose stack, standard for VPS-based solo stakers
• Stereum — GUI-driven setup with client diversity options
• Rocket Pool node — if you want to run a megapool validator (see Method 2)

Yield in 2026: ~3.0–3.3% gross with no protocol fees. You capture 100% of MEV + priority fees. The marginal yield over liquid staking is ~0.6–1.0 percentage points — meaningful at scale, negligible at 32 ETH.

Best for: Holders with 32+ ETH and willingness to operate. The marginal contribution to network decentralization is concrete; the marginal yield premium is real but modest.

Method 2: Liquid staking (recommended for most)

Liquid staking (depositing ETH to a protocol that returns a transferable token representing the staked position) deposits ETH to Lido, Rocket Pool, or Ether.fi and returns a liquid token (stETH, rETH, eETH) that accrues yield and works as DeFi collateral. Last verified: 2026-05-27.

You deposit ETH; you receive a liquid token (stETH from Lido (Lido is the largest Ethereum liquid staking protocol, issuing stETH to depositors), rETH from Rocket Pool (Rocket Pool is a permissionless decentralized liquid staking protocol with node operators bonding as little as 4 ETH per megapool validator after the Saturn I upgrade), eETH from Ether.fi (Ether.fi is a non-custodial liquid restaking protocol where depositors retain their validator keys and earn both staking and EigenLayer AVS rewards)) that accrues staking yield and can be used in DeFi.

Lido (stETH)

• Market share: ~24% of all staked ETH (~8.72M ETH, ~$19B TVL) — down from a ~32% peak in 2023 due to LST and LRT competition
• How it works: Deposit ETH to Lido's contract; receive stETH 1:1. Lido distributes ETH across ~40 professional node operators (Figment, Kiln, P2P, Stakefish, Chorus One, etc.), governed by Lido DAO with LDO token holders selecting operators. Daily rebasing increases your stETH balance automatically.
• Yield (May 2026): ~2.4% APY net of Lido's 10% performance fee (5% to node operators, 5% to Lido DAO treasury), per vaults.fyi 7-day average
• Withdrawals: Instant via Curve stETH/ETH pool (typically within 10–30 bps of peg), or queued direct unstake through the Lido withdrawal contract (1–5 days at exact 1:1 redemption)
• Trade-offs: Most centralized of the major LSTs — Lido DAO controls operator selection. Concentration risk: if Lido stake exceeds 33% of total Ethereum stake, it would threaten consensus finality. Lido has self-limited; market share has declined naturally. The Simple DVT module has added permissionless operators since 2024.

Rocket Pool (rETH)

• Market share: ~3% (smaller than Lido but the leading permissionless LST)
• How it works after Saturn I (Feb 2026): Deposit ETH; receive rETH whose exchange rate appreciates over time (price-appreciating, not rebasing — better for DeFi composability and tax tracking in most jurisdictions). Node operators now bond as little as 4 ETH per megapool validator (down from 8 ETH pre-Saturn), doubling capital efficiency and driving a 35% surge in active nodes during Q1 2026. RPL's role shifted: the fee switch is now active, making RPL an ETH-accrual token rather than purely inflationary collateral.
• Yield (May 2026): ~2.0% APY per vaults.fyi 7-day average (slightly lower than stETH due to the commission structure and current operator set size)
• Withdrawals: Instant via DEX (Uniswap, Balancer rETH/ETH pools) or queued via the Rocket Pool deposit pool (1–5 days)
• Trade-offs: Smaller liquidity and TVL than stETH; fewer integrations. rETH is accepted as collateral on Aave V3 and Morpho but with lower LTV caps than stETH.

Ether.fi (eETH / weETH)

• Market share: ~$7.8B TVL (May 2026), the largest liquid restaking token (LRT) by TVL
• How it works: Non-custodial — depositors retain validator keys via a NodeOperators-as-a-Service model where Ether.fi infrastructure runs the validator but the keystore is split. Native EigenLayer integration: every staked ETH is automatically restaked, earning both Ethereum staking yield and EigenLayer AVS rewards. weETH is the wrapped non-rebasing version for DeFi composability.
• Yield (May 2026): ~2.6% base staking (per vaults.fyi), plus EigenLayer AVS rewards on top — total restaking APY ranges 3.8–6% depending on AVS selection
• Withdrawals: Queued unstake or DEX swap (weETH on Uniswap, Curve, Balancer)
• Trade-offs: Newer protocol (launched 2023). Restaking exposure means slashing surface includes EigenLayer AVSs in addition to Ethereum's base layer. EigenLayer's slashing logic has been live since late 2025 — this is a real additional risk.

For most users: Lido stETH (most liquid, most DeFi integrations) or Rocket Pool rETH (most decentralized, permissionless). For yield-seeking users comfortable with restaking risk: Ether.fi eETH.

Method 3: CEX staking

CEX staking is one click on Coinbase, Kraken, or Binance — nets 2.1–2.6% after the exchange takes a 25–30% cut, with 5–14 day unstake delays and full counterparty risk. Last verified: 2026-05-27.

Click "stake ETH" on Coinbase, Kraken, or Binance. They run validators on your behalf and pay you after taking a 25–30% cut. Kraken advertises up to 2.57% net APY on ETH (their page, May 2026) and charges 26% on bonded staking. Coinbase charges 25%, which on a ~2.84% consensus yield delivers roughly 2.1–2.2% net. Binance varies by region and KYC tier.

Pros: Zero setup, no separate tokens to manage, no DeFi UX.

Cons: Lowest net yield (2.1–2.6%); counterparty risk (FTX, Celsius, BlockFi precedents); illiquidity during unstake (5–14 days standard, longer if the validator exit queue extends); regulatory exposure (Kraken's 2023 SEC settlement forced US users off its staking product; Coinbase's regulatory position stabilized after the 2024–2025 SEC reset under new leadership, but custodial staking-as-a-service remains a policy target in some jurisdictions).

Use only if: You already trust the exchange with the ETH, the amount is modest, and zero operational complexity is the priority.

Method 4: DVT (Distributed Validator Technology)

DVT (Distributed Validator Technology splits a single Ethereum validator across multiple independent operators using threshold signatures for fault tolerance) via SSV Network or Obol splits a validator across multiple operators. One can fail without slashing the whole stake. Last verified: 2026-05-27.

SSV Network and Obol let you split a validator across multiple independent operators using threshold signatures. If one operator fails (loses internet, crashes, is seized), the others continue producing valid attestations and block proposals. The validator stays online; the stake stays safe. This eliminates single-operator slashing risk from key-management failures and dramatically reduces correlated downtime.

Both Lido (via the Simple DVT module) and Ether.fi use DVT operationally in some configurations. Direct DVT staking is most relevant for: solo stakers who want resilience without running multiple machines themselves; institutional stakers wanting fault tolerance for treasury operations; or DAOs splitting a community validator across trusted members.

Yields match solo staking (~3.0–3.3%) net of small DVT operator fees. Setup is more involved than standard solo staking — SSV and Obol both require their own operator-selection and key-split workflows.

Where to deploy your LST after staking

LSTs unlock DeFi composability: lend on Aave (+0.5–1%), buy fixed yield on Pendle (5–8%), LP on Curve (2–5% + CRV), or restake via EigenLayer (total 3.8–6%). Last verified: 2026-05-27.

Liquid staking tokens earn yield and unlock DeFi composability — the primary reason most users choose them over solo staking:

→ Find the right app with web3wagmi Atlas (pick what you want to do with your LST — stake, lend, LP, restake — and Atlas shows the protocols that do it)

Tip: Most cost-efficient passive setup in 2026 is eETH on Ether.fi or weETH used as collateral on Aave — you get base staking + EigenLayer restaking + lending utility in one position. The looped variant (deposit weETH, borrow ETH, restake) can push effective yields toward 6–8% but introduces liquidation risk on any LST depeg event.

Best ETH staking method by use case

Lido stETH for under 32 ETH; solo staking for max decentralization; Rocket Pool for trust-minimised LST; Ether.fi eETH for restaking; Coinbase for one-click simplicity. Last verified: 2026-05-27.

• Best ETH staking method for HODLers under 32 ETH — Lido stETH (most liquid, ~24% market share, ~$19B TVL, most DeFi integrations).
• Best ETH staking method for max decentralization — Solo staking (32 ETH + DappNode appliance + minority client pair).
• Best ETH staking method for decentralization without solo ops — Rocket Pool rETH (permissionless node operators, post-Saturn I megapool design).
• Best ETH staking method for DeFi composability — stETH on Lido (most-accepted LST collateral on Aave, Morpho, Pendle).
• Best ETH staking method for restaking exposure — Ether.fi eETH (native EigenLayer integration, $7.8B TVL, largest LRT).
• Best ETH staking method for fixed yield — Pendle PT-stETH or PT-eETH (lock yield for 3–12 months at maturity).
• Best ETH staking method for simplicity — Coinbase staking (one click; surrenders ~25% of yield to fees).
• Best ETH staking method for 4–32 ETH node operators — Rocket Pool megapool (4 ETH bond per validator post-Saturn I + RPL revenue share).
• Best ETH staking method for institutional treasury — Lido or direct partnership with Figment / Kiln / Allnodes with DVT overlay.
• Best ETH staking method to avoid — Any LST under 12 months old with under $100M TVL — depeg risk is disproportionate in stress conditions.

Step-by-step walkthrough: Lido in 10 minutes

The path of least resistance for most ETH holders. Total time: under 10 minutes assuming wallet already funded.

1. Bookmark the official URL — stake.lido.fi. Type it; don't click search results. Lido phishing sites have been a recurring problem since 2023.
2. Connect your wallet. MetaMask, Rabby (preferred — signature simulation catches approval traps), or a hardware wallet via Ledger/Trezor passthrough.
3. Enter the ETH amount. No minimum. Leave at least 0.01 ETH for gas; stake the rest.
4. Review the transaction. Rabby will show: you send X ETH, you receive X stETH (1:1 at deposit time). Confirm the contract address against docs.lido.fi.
5. Sign and submit. Gas typically runs $1–5 in 2026 at normal network load. The deposit routes to Lido's contract, which allocates to its node operator set.
6. stETH appears in your wallet after one block (~12 seconds).
7. Daily rebasing kicks in automatically. Your stETH balance increases ~0.0066% per day at current ~2.4% APY. No action needed.
8. Optional: deploy in DeFi. Aave V3 (collateral for ETH borrowing in e-mode), Curve (LP for trading fees), Pendle (fixed-rate PT sale), Morpho (lending).
9. Optional: unstake when needed. Curve stETH→ETH swap is instant, usually within 10–30 bps of peg. Direct unstake via Lido's withdrawal page takes 1–5 days but redeems at exact 1:1.

Risks summary

Five real risks: smart-contract bugs, LST depeg (stETH hit $0.93 in 2022), slashing, regulatory action against custodial staking, and compounding restaking exposure across EigenLayer AVSs. Last verified: 2026-05-27.

• Smart contract risk. Every LST is a smart contract. Lido has been audited by Statemind, Sigma Prime, Quantstamp, Hexens, and ChainSecurity and has paid out bug bounties since 2021. Rocket Pool similarly. Ether.fi is newer with proportionally fewer audit-years under its belt. No exploits to date for any of the top three, but "no exploits yet" is the historical posture of every protocol before they have one.
• LST depeg risk. stETH dropped to $0.93 in June 2022 during the Celsius and 3AC unwinds — both held large stETH positions that were force-sold into thin Curve liquidity. Recovery took weeks. rETH held near peg in the same period due to tighter arbitrage mechanics. Assume LSTs can wobble in stress: if you are using leverage against stETH on Aave, set your health factor conservatively (1.5+).
• Slashing risk. The historical slashing rate across Lido's operator set is well under 0.05% of validator-years. The maximum protocol-level slash is 1 ETH initial penalty plus correlated exit penalties (theoretically up to the entire stake but practically far smaller). Normal downtime costs missed-attestation rewards (~0.01 ETH/year for poor uptime) and the inactivity-leak quadratic penalty only if you're offline during a finality-stall event.
• Regulatory risk. The SEC's 2023 Kraken settlement ($30M, forced US exit from staking-as-a-service) remains a precedent. Coinbase's regulatory position has stabilized after the 2024–2025 SEC reset, and the 2024 SEC guidance suggested custodial staking products differ from decentralized protocols. Self-staking and permissionless LSTs (Lido, Rocket Pool) sit in a more defensible position but are not immune.
• Restaking compounding risk. EigenLayer slashing went live in stages through 2025 — slashing logic is active but large real-world penalty events have not yet triggered as of May 2026. Exposure stacks: eETH on Aave + EigenLayer restaking + active AVS opt-ins = four distinct slashing surfaces in one position. The first material AVS slashing event will be a pricing event for LRTs.

Looking ahead

Four signals to track over the next 12 months:

• ETF staking distributions. Grayscale's ETHE distributed its first staking reward ($0.083/share) on January 6, 2026. BlackRock and Fidelity ETF staking approvals are expected in 2026 H2, which will add tens of billions in staking demand and compress consensus yields toward the 2.4–2.7% range.
• Pectra validator consolidation. With MaxEB at 2,048 ETH, large stakers (Coinbase, Lido, Kiln) are collapsing thousands of 32-ETH validators into fewer large-balance validators, reducing attestation traffic and beacon-chain load. Individual yields are marginally unaffected but network scalability improves meaningfully.
• DVT adoption growth. Lido's Simple DVT module and Stakewise V3 DVT implementations are expanding. By end of 2026, DVT-backed stake is on track to exceed 10% of the total. This reduces correlated-failure risk systemically.
• First EigenLayer AVS slashing event. EigenLayer slashing logic is live; real penalties on a misbehaving AVS have not yet triggered. When the first event occurs, expect brief LRT depeg volatility — a liquidity test for the restaking sector.

Verdict by user type

HODLers under 32 ETH should use Lido or Rocket Pool. Solo stakers want 32+ ETH and ops capacity. DeFi users want eETH or stETH. Yield maxers stack eETH + EigenLayer + Pendle. Last verified: 2026-05-27.

• HODLer with under 32 ETH: Lido stETH or Rocket Pool rETH. Set and forget.
• HODLer with 32+ ETH: Solo staking if you have ops capacity (note ~57-day activation queue in May 2026); Rocket Pool megapool if not.
• DeFi user: Ether.fi eETH or Lido stETH, deploy as collateral on Aave V3 / Morpho.
• Yield maxer: eETH on Ether.fi + EigenLayer AVS restaking + Pendle PT-eETH fixed-rate exit at maturity.
• Maximum simplicity: Coinbase or Kraken staking. You'll surrender ~25–30% of yield but it takes one click and no wallet management.

---

Related: DeFi Yield Farming Ultimate Guide