On this page
- What is the Pectra upgrade and when did it go live?
- What does EIP-7702 change for regular Ethereum accounts?
- What does EIP-7251 change for Ethereum staking?
- What does EIP-7691 change for blobs and layer-2 fees?
- What other EIPs shipped in Pectra?
- What happened after Pectra: Fusaka and PeerDAS?
- Ethereum upgrade timeline since The Merge
- What should users do now?
- FAQ
- Source register
On 7 May 2025 at 10:05 UTC, Ethereum activated the Pectra upgrade at epoch 364032, combining the Prague execution-layer and Electra consensus-layer changes into a single hard fork of 11 EIPs. Pectra raised the staking balance cap from 32 ETH to 2,048 ETH, doubled the blob target for layer-2 data, and let ordinary accounts act like smart contracts. For most users nothing changed on the day itself; the effects surface through wallet and staking features that are still rolling out as of November 2026.
This explainer covers each headline change, the new EIP-7702 phishing risk, and where Pectra sits in Ethereum’s upgrade roadmap. For background on how the network and its native asset work, see the main coin guides hub.
What is the Pectra upgrade and when did it go live?
Pectra is a combined Ethereum hard fork that went live on mainnet at epoch 364032 on 7 May 2025 at 10:05 UTC. It merges the Prague execution-layer and Electra consensus-layer changes into 11 EIPs. Its headline changes are smart accounts (EIP-7702), a higher staking balance cap (EIP-7251) and more blob capacity for layer-2 networks (EIP-7691).
The name “Pectra” is a portmanteau of Prague and Electra, the names of the execution-layer and consensus-layer halves of the fork, as listed on ethereum.org’s Pectra page. Pectra followed the Dencun upgrade of 13 March 2024 and was in turn followed by Fusaka on 3 December 2025. The upgrade did not create a new coin or alter ETH supply: it changed how accounts, staking and block data work on the existing chain.
What does EIP-7702 change for regular Ethereum accounts?
EIP-7702 lets a normal externally owned account (EOA), the kind created from a seed phrase, authorise its address to run smart-contract code during a transaction. This enables transaction batching, sponsored gas, social recovery and spending limits without moving funds to a new account. The trade-off is new risk: signing a malicious authorisation can hand control of the account to attacker code.
An EOA is the standard account type controlled by a private key. Before Pectra, an EOA could only send simple transfers or call contracts; it could not hold its own logic. EIP-7702 adds a new transaction type, 0x04, that lets an EOA set a delegation pointer to smart-contract code. The pointer is written as 0xef0100 followed by the contract address, so the account behaves like that contract without permanently changing its key, according to the EIP-7702 specification.
The user-facing benefits, per ethereum.org, include:
- Batching several transactions into one, saving repeated approvals and gas
- Sponsored (gasless) transactions, where an app or friend pays the fee
- Spending limits and social recovery, so losing a key does not mean losing the account
- Passkey and biometric sign-in instead of a seed phrase for daily use
The phishing risk is real. The EIP-7702 spec states that the delegated code has unrestricted access to the account and that there is no safe way to offer a raw authorisation-signing interface, so wallets are told not to expose one. A signed authorisation is effectively a grant of control over the account. Attackers have used this to trick people into signing a message that points their account at malicious code. If you use a wallet such as MetaMask, only ever act on prompts the wallet itself generates for a transaction, and never sign a standalone “authorisation” message from a stranger or an unknown site. Treat any request to sign a message you cannot read as a phishing attempt, the same rule covered in our seed phrase safety guide.
Delegation is also one-use and revocable. Each authorisation burns a nonce, so a signed authorisation cannot be replayed, and sending an authorisation that points at the zero address clears the delegation. A wallet can also cancel a pending authorisation by using up the account nonce. The chain ID in the authorisation matters too: an authorisation with chain ID 0 is valid on every chain, which widens the blast radius of a bad signature, so prefer delegations scoped to one chain.
What does EIP-7251 change for Ethereum staking?
EIP-7251 raises the validator maximum effective balance from 32 ETH to 2,048 ETH, so a validator can stake anywhere from 32 to 2,048 ETH and earn rewards on the amount above 32. It also adds consolidation, letting operators merge up to 64 validators into one, plus a new compounding withdrawal credential type. The 32 ETH activation minimum is unchanged.
Under proof-of-stake, each validator had an effective balance capped at 32 ETH, so a large staker ran many separate validator keys. The EIP-7251 specification raises the MAX_EFFECTIVE_BALANCE constant to 2,048 ETH while keeping the minimum activation balance at 32 ETH. Rewards now accrue in 1 ETH steps above 32: a 33 ETH validator earns on the extra 1 ETH, as ethereum.org explains. The effective balance only moves up once the actual balance passes the next step by 0.25 ETH, a detail documented by Consensys.
| Feature | Before Pectra | After Pectra |
|---|---|---|
| Maximum effective balance | 32 ETH | 2,048 ETH |
| Rewards on balance above 32 ETH | Paid out as partial withdrawals | Can compound on the validator via 0x02 credentials |
| Consolidating validators | Not available | Up to 64 validators merged into one |
| Minimum to activate | 32 ETH | 32 ETH (unchanged) |
Worked example: a solo staker running 64 validators of 32 ETH each, or 2,048 ETH in total, can now consolidate them into a single 2,048 ETH validator. This cuts the number of validator keys, signatures and beacon-chain records to maintain, without changing the staked amount or the 32 ETH floor. The figure of up to 64 validators per consolidation is reported by Consensys, and the 2,048 ETH ceiling comes from the EIP-7251 spec.
For everyday stakers, most of this is handled by the staking service or pool they use, so no manual action is usually needed. See our how to stake Ethereum guide for the mechanics, and liquid staking explained if you hold a token such as stETH instead of running a validator.
What does EIP-7691 change for blobs and layer-2 fees?
EIP-7691 raises Ethereum’s blob target from 3 to 6 blobs per block and the maximum from 6 to 9, roughly doubling the data space available to layer-2 rollups such as Arbitrum and Base. More data capacity lowers what rollups pay Ethereum, which is passed on to users as lower fees. It also changes how quickly the blob base fee adjusts.
Blobs are the cheap data sidecar introduced by EIP-4844 in the Dencun upgrade of March 2024, and they are what layer-2 networks post their transaction data to. The EIP-7691 specification raises the consensus-layer constants to a target of 6 and a maximum of 9 blobs per block, up from 3 and 6. This changes the target-to-maximum ratio from 1:2 to 2:3.
| Upgrade | Activation | Blob target per block | Blob maximum per block |
|---|---|---|---|
| Dencun (EIP-4844) | 13 March 2024 | 3 | 6 |
| Pectra (EIP-7691) | 7 May 2025 | 6 | 9 |
The blob base fee now adjusts more smoothly: when blocks are full the fee rises about 8.2% per adjustment (previously about 12.5%), and when blocks are empty it falls about 14.5% (previously about 11.1%). Those figures are stated in the EIP-7691 spec and describe the pricing mechanism, not a current fee, so no live gas price is quoted here. The practical effect for users is that layer-2 networks gained headroom, which should reduce data-driven fee spikes, though actual rollup fees also depend on each network’s own demand and its operator’s pricing.
What other EIPs shipped in Pectra?
Beyond the three headline changes, Pectra shipped eight further EIPs covering validator exits, deposits, calldata pricing, cryptography and attestations. EIP-7002 lets withdrawals be triggered with the withdrawal credentials, EIP-6110 moves deposits on-chain, and EIP-7549 cuts the BLS signatures needed per epoch by a factor of 64. These mostly affect validators, staking services and developers rather than everyday users.
| EIP | What it does |
|---|---|
| EIP-7002 | Lets withdrawal credentials trigger a validator exit from the execution layer |
| EIP-6110 | Processes new-validator deposits on-chain for faster activation |
| EIP-7623 | Raises calldata pricing only for data-heavy transactions, leaving over 99% unaffected |
| EIP-7549 | Moves the committee index out of the attestation, cutting BLS signatures per epoch by a factor of 64 |
| EIP-2537 | Adds BLS12-381 curve precompiles |
| EIP-2935 | System contract storing the last 8,192 block hashes |
| EIP-7685 | General-purpose execution-layer requests, used for consolidations and exits |
| EIP-7840 | Adds a blob schedule to execution-layer client configuration |
The first six rows above are summarised from ethereum.org’s Pectra page; the 64-fold BLS signature reduction for EIP-7549 is quantified by Consensys. EIP-7685 is the request framework that EIP-7251 and EIP-7002 build on.
What happened after Pectra: Fusaka and PeerDAS?
The next upgrade, Fusaka (Fulu-Osaka), activated on 3 December 2025 at 21:49 UTC at block 23,935,694 and epoch 411,392. Its headline change is PeerDAS (EIP-7594), which lets nodes verify blob data by sampling a fraction of it instead of downloading everything, and it raised the default block gas limit to 60 million. As of November 2026 the next major upgrade, Glamsterdam, is scheduled for Q4 2026.
PeerDAS, formally EIP-7594, extends blob data with erasure coding and splits it into cells, so a node only samples and custodies a subset of columns rather than the whole blob. Ethereum.org’s Pectra page notes that PeerDAS shipped in Fusaka and allows higher blob counts still. The ethereum.org roadmap records Fusaka’s default block gas limit being raised to 60 million and describes follow-up “blob parameter only” (BPO) forks as a way to raise blob counts between major upgrades without a full hard fork.
This is the dated status check most older Pectra explainers lack: as of November 2026, Fusaka is live on mainnet, PeerDAS is active, and the next major fork, Glamsterdam, is scheduled for Q4 2026 with enshrined proposer-builder separation and block-level access lists, with scope still not final, per the ethereum.org roadmap. A further upgrade, Hegotá, is slated for 2027.
Ethereum upgrade timeline since The Merge
Since replacing proof-of-work with proof-of-stake in the 15 September 2022 Merge, Ethereum has shipped four further mainnet upgrades. Shapella enabled staking withdrawals (12 April 2023), Dencun introduced blobs (13 March 2024), Pectra added smart accounts and a bigger staking cap (7 May 2025), and Fusaka brought PeerDAS (3 December 2025).
| Date | Event | Source |
|---|---|---|
| 15 September 2022 | Paris / The Merge: consensus moves from proof-of-work to proof-of-stake | Ethereum.org |
| 12 April 2023 | Shanghai-Capella / Shapella: staking withdrawals enabled | Ethereum.org |
| 13 March 2024 | Cancun-Deneb / Dencun: EIP-4844 data blobs for layer-2 networks | Ethereum.org |
| 7 May 2025 | Prague-Electra / Pectra: EIP-7702, EIP-7251, EIP-7691 and eight more EIPs | Ethereum.org |
| 3 December 2025 | Fulu-Osaka / Fusaka: PeerDAS (EIP-7594) and a 60 million gas limit | Ethereum.org |
All five dates and event names above come from the ethereum.org history page, which records The Merge at 15 September 2022, Shapella at 12 April 2023, Dencun at 13 March 2024, Pectra at 7 May 2025 and Fusaka at 3 December 2025. The Merge and Dencun are also covered on our Ethereum explainer.
What should users do now?
For most holders nothing changed on 7 May 2025: your ETH, wallets and keys kept working and no action was required. The practical steps are the same as always: update wallet and staking software when prompted, treat new EIP-7702 authorisation prompts with caution, and never sign a message you do not understand. Staking-service users should confirm whether their provider supports the new features.
- Update your software. Wallet apps, browser extensions and validator clients release Pectra- and Fusaka-compatible versions; running the latest version from the official source keeps you on the correct chain.
- Be careful with EIP-7702 prompts. If a site asks you to sign a message that sets account code or “authorises” a contract, stop. Legitimate wallets surface these as ordinary transaction reviews, not as blind message signing.
- Check your staking setup. If you stake through a service, confirm whether it has adopted EIP-7251 consolidation and 0x02 compounding withdrawals, and only change credentials through the official interface.
- Keep custody basics in place. Never share a seed phrase, and store backups as described in our seed phrase guide. New account features do not change the rule that whoever holds the key or the signed authorisation controls the funds.
If you are holding or trading ETH, treat the upgrade as an infrastructure change rather than a price signal. Crypto assets are volatile and this is not financial advice. For choosing where to buy or keep ETH, our wallet guides and exchange guides compare options on fees and safety, not on price outlook.
FAQ
When did the Ethereum Pectra upgrade go live?
Pectra went live on 7 May 2025 at 10:05 UTC at epoch 364032. It combined the Prague execution-layer and Electra consensus-layer changes into 11 EIPs.
Does Ethereum still use proof-of-work?
No. Ethereum replaced proof-of-work with proof-of-stake in The Merge on 15 September 2022. Blocks are now proposed and attested by validators that stake ETH, not by miners running hardware.
Is Ethereum 2.0 out yet?
The term Ethereum 2.0 was retired by the Ethereum community. The features it described now ship as regular upgrades: The Merge (2022), Shapella (2023), Dencun (2024), Pectra (2025) and Fusaka (2025).
What does EIP-7702 mean for my wallet security?
EIP-7702 lets an ordinary account delegate to smart-contract code, which enables batching and recovery, but a malicious authorisation could give attacker code control of the account. Never sign a raw authorisation message you do not understand, and use a wallet that shows which code an account is delegating to.
What is the Glamsterdam upgrade?
Glamsterdam is the next major Ethereum upgrade after Fusaka, scheduled for Q4 2026. As of November 2026 its planned scope includes enshrined proposer-builder separation and block-level access lists, but the scope is not final.
What is causing Ethereum to rise?
Price moves are driven by market demand and are not caused by any single protocol upgrade. Pectra and Fusaka changed network capacity and staking mechanics, not the supply schedule or a price level. Crypto assets are volatile and this is not financial advice.
Source register
- 01Ethereum.org: Prague-Electra (Pectra)ethereum.org
- 02Ethereum.org: Ethereum historyethereum.org
- 03Ethereum.org: Ethereum roadmapethereum.org
- 04Ethereum EIPs: EIP-7702eips.ethereum.org
- 05Ethereum EIPs: EIP-7251eips.ethereum.org
- 06Ethereum EIPs: EIP-7691eips.ethereum.org
- 07Ethereum EIPs: EIP-7594eips.ethereum.org
- 08Consensys: Understanding Ethereum's Pectra upgradeconsensys.com
Last updated Spotted an error? Send a correction
