What Was the Pectra Upgrade?

Pectra was Ethereum's first major hard fork after Dencun, merging two coordinated tracks into one release. The name fuses Prague, the execution-layer half, with Electra, the consensus-layer half, following Ethereum's habit of pairing a Devcon host city with a star name.

By EIP count it was the largest upgrade since the 2022 Merge, carrying eleven proposals where Dencun shipped six. The scope was deliberately broad rather than deep on any single theme, spreading work across wallet design, validator economics and data availability.

We tracked the rollout through a rough testing cycle. Early activations on Holesky and Sepolia hit finality problems in February 2025, which pushed developers to spin up a fresh testnet called Hoodi as a final dry run. Once Hoodi held, the team locked mainnet activation for epoch 364032 on 7 May 2025.

The Eleven Pectra EIPs That Shipped

Pectra's proposals cluster into three jobs: making accounts programmable, improving the staking experience, and giving rollups more room. Grouping them that way makes the upgrade easier to read than the raw list.

Account and wallet changes

• EIP-7702: Lets an externally owned account temporarily adopt smart-contract code by signing a delegation to a chosen contract. This is the headline feature, enabling transaction batching, gas sponsorship, spending limits and passkey-style authentication without migrating to a new address.
• EIP-2537: Adds a precompile for BLS12-381 curve operations, lowering the cost of the cryptography that ZK rollups and staking protocols lean on.

Validator and staking changes

• EIP-7251: Raises the maximum effective balance from 32 ETH to 2,048 ETH for validators that opt into the new compounding credentials, while keeping the 32 ETH minimum to join.
• EIP-6110: Moves validator deposits onto the execution layer, cutting the wait between a deposit and activation from roughly nine hours to about thirteen minutes.
• EIP-7002: Allows withdrawals and exits to be triggered from a validator's withdrawal address, so the owner of the funds can always exit without relying on the signing key operator.
• EIP-7549: Moves the committee index outside the attestation structure, reducing the data validators process and improving consensus efficiency.

Scaling and infrastructure changes

• EIP-7691: Doubles blob throughput, lifting the target from three to six blobs per block and the maximum from six to nine, which gives Layer 2 networks more space to post compressed data.
• EIP-7623: Increases the cost of calldata to nudge rollups towards blobs and to cap the worst-case size of a block, offsetting the bandwidth that the blob increase adds.
• EIP-2935: Saves historical block hashes directly in state, useful groundwork for stateless clients and cross-chain proofs.
• EIP-7685: Creates a general framework for the execution layer to pass requests to the consensus layer, the plumbing that EIP-6110 and EIP-7002 rely on.
• EIP-7840: Adds a blob schedule to client config files, letting future blob-capacity changes ship without a full hard fork.

Several proposals once pencilled in for Pectra, including PeerDAS and the EVM Object Format, were held back. Most of those landed later in Fusaka or remain on the roadmap.

Pectra One Year On: How It Actually Played Out

Judged against its own goals, Pectra delivered the technical work cleanly and then ran into the messier reality of how people use new primitives. The activation itself was smooth, with no consensus incidents at the fork, which set it apart from the bumpy testnet phase.

Adoption of the marquee feature was immediate. Within a week of launch, more than 11,000 EIP-7702 authorisations were recorded on mainnet, with exchanges such as WhiteBIT and OKX leading early integration. On the validator side, the share of staked ETH sitting in consolidated, compounding validators climbed from roughly 2% to over 11% in about six months.

The honest reading is that Pectra succeeded as infrastructure while producing two surprises. The first was how quickly attackers weaponised account abstraction, which we cover below. The second was the market reaction, which ran hotter at launch than almost anyone predicted before fading over the following year.

EIP-7702 and the Smart-Account Era

EIP-7702 reframed what an Ethereum wallet can be. Rather than forcing users onto a separate smart-contract wallet, it lets a normal address point to contract code for the duration of a delegation, then revoke or replace it at will. Three years of history, NFTs and DeFi positions stay attached to the same address.

The legitimate use cases arrived fast. Circle built gasless USDC transfers on top of it, wallets started bundling approve-and-swap into one signature, and session keys made repeated app interactions far less tedious. For anyone who has explained gas to a newcomer, this was the most user-facing change Ethereum had shipped in years.

The trouble surfaced within weeks. By the end of May 2025, security firm Wintermute found that more than 97% of EIP-7702 delegations on mainnet pointed to near-identical wallet-draining contracts, the most reused of which it nicknamed "CrimeEnjoyor." One victim lost close to $150,000 in a single batched transaction tied to the Inferno Drainer kit.

The nuance matters here. These sweepers do not exploit a flaw in EIP-7702 itself, since they only work against wallets whose private keys are already compromised through phishing. What the feature changed is the speed and efficiency of theft once a key leaks, which is why wallet providers raced to display delegation targets and flag suspicious authorisations. If you hold assets in a MetaMask-style wallet, treating any delegation request with the same caution as exporting a seed phrase is the practical takeaway.

What Pectra Changed for Stakers

For people running validators, EIP-7251 was the proposal with the most day-to-day weight. Before Pectra, every validator was capped at 32 ETH of effective balance, so rewards above that ceiling sat idle until they could fund a whole new 32 ETH validator. Large operators ended up running hundreds of redundant validators that bloated the network.

The 2,048 ETH ceiling fixed both problems. Operators can now consolidate many validators into one and earn rewards on the full balance, with auto-compounding adding roughly a 1% to 1.5% relative uplift in staking yield. Pectra also cut the initial slashing penalty sharply, which lowered the risk of holding a large stake in a single validator.

There is a centralisation tension built into this. Consolidation is good for efficiency and bandwidth, yet concentrating more stake behind fewer keys raises the stakes of any single failure. For most solo stakers the maths still favours staying small, while professional operators with strong slashing protection and distributed validator setups gained the most. Our staking calculator is a useful way to model what consolidation would mean for a given balance.

How ETH Price Reacted to Pectra

The price story is where hindsight rewrites the pre-launch narrative most sharply. Heading into the upgrade, Ethereum had been the laggard of the major assets, sliding to a two-year low near $1,417 in early April 2025 and underperforming the broader market badly through the first quarter.

Then the upgrade flipped the mood. ETH jumped more than 20% in the two days around activation, its largest single-day gain since May 2021, helped along by a short squeeze and a risk-on macro backdrop. From there the rally extended through the summer, carrying ETH from around $1,800 to an all-time high near $5,000 in August 2025.

Crediting all of that to Pectra would overstate the case. Spot ETH ETF inflows, easing trade tensions and broad altcoin rotation did much of the lifting, and exchanges logged heavy profit-taking as the price climbed. A year later ETH trades back near $1,785, a reminder that protocol upgrades shape the network's long-term capability far more reliably than its short-term price.

Where Pectra Sits in the Roadmap

Pectra was never meant to be a finish line. It cleared technical debt and shipped user-facing wins while leaving the heavier scaling work for the forks that followed.

Fusaka arrived on 3 December 2025, built around PeerDAS, the data-availability sampling technique that lets nodes verify rollup blobs without downloading every one. It picked up much of what Pectra deferred and pushed L2 data costs down further.

Next on the schedule is Glamsterdam, targeting the second half of 2026. Its headline proposals, enshrined proposer-builder separation and block-level access lists, aim to bring block building on-chain and unlock parallel execution, continuing the throughput push that Pectra and Fusaka began.

Final Thoughts

Seen from a year out, Pectra reads as a competent, unglamorous upgrade that did most of what it promised. It made Ethereum accounts programmable, gave stakers a far better deal, and handed rollups more room to grow, all without a single mainnet hiccup at the fork.

The wrinkles are just as instructive. The CrimeEnjoyor episode showed that powerful new primitives invite abuse the moment they ship, and the price round-trip showed how little upgrades have to do with where ETH trades on any given day. For builders and stakers, though, the foundation Pectra laid is the part that endures, and the forks since have built directly on top of it.