Ethereum's next major upgrade targets a 200 million gas limit, enshrined proposer-builder separation via EIP-7732, and block-level access lists under EIP-7928, with a tentative mainnet activation in June 2026.
Ethereum's core developers have finalised the scope of Glamsterdam, the network's next hard fork and its most structurally ambitious upgrade since the Merge. The upgrade centres on two headline proposals — enshrined proposer-builder separation under EIP-7732 and block-level access lists under EIP-7928 — that together promise to reshape how blocks are constructed, validated, and executed on the world's largest smart contract platform.
Vitalik Buterin outlined eight Ethereum Improvement Proposals in late February that define the upgrade's technical boundaries. The most consequential changes include raising the gas limit from 60 million to 200 million per block, enabling throughput of approximately 10,000 transactions per second, and reducing gas fees by an estimated 78.6% across both simple transfers and complex smart contract interactions. Glamsterdam carries a tentative mainnet activation target of June 2026, though developers have stressed that the timeline remains aspirational pending completion of public testnet phases and dual security audits.
Enshrined Proposer-Builder Separation
EIP-7732 represents one of the most significant changes to Ethereum's block production pipeline since the transition to proof of stake. Under the current system, approximately 80 to 90 percent of Ethereum blocks are constructed through MEV-Boost, a third-party relay system that sits outside the protocol itself. Block builders assemble transaction bundles optimised for maximal extractable value, relay operators verify and transmit those bundles, and proposers select the highest-paying block to include in the chain.
This architecture has drawn persistent criticism for concentrating power in a small number of relay operators. Flashbots, the dominant relay provider, has processed more than 90% of MEV-Boost blocks at various points, creating a centralisation bottleneck that contradicts Ethereum's decentralisation ethos. The collapse of a relay operator or a targeted attack on relay infrastructure could disrupt block production across the entire network.
EIP-7732 eliminates this dependency by moving proposer-builder separation directly into the Ethereum protocol. Under the new design, builders assemble blocks and cryptographically seal their contents before submitting them to proposers. Proposers select the highest-paying block without being able to see or modify the transactions inside, and the block's contents are only revealed after finalisation. This mechanism preserves the competitive dynamics of the current builder market while removing the need for trusted intermediaries.
Block-Level Access Lists and Parallel Execution
EIP-7928 introduces block-level access lists, a mechanism that enables the Ethereum Virtual Machine to process transactions in parallel rather than sequentially. Under the current architecture, every transaction in a block is executed one after another, even when the transactions interact with entirely different contracts and storage slots. This serial execution model leaves significant computational capacity unused and represents one of the primary bottlenecks constraining Ethereum's throughput.
Block-level access lists require each transaction to declare upfront which accounts and storage slots it will read from or write to. This declaration is committed in the block header and verified against the actual execution trace after the block is processed. When the EVM encounters transactions that access non-overlapping state, it can execute them simultaneously across multiple processing threads. Benchmarks conducted on Ethereum's Holesky testnet have demonstrated throughput improvements of 3 to 5 times compared to serial execution, with the precise gain depending on the composition of transactions within each block.
The interaction between ePBS and BALs introduces engineering complexity that has not been tested at mainnet scale. Builders must now optimise not only for MEV extraction but also for parallelism, grouping transactions with non-overlapping state access to maximise the throughput benefits of EIP-7928. Ethereum Foundation researcher Dankrad Feist has acknowledged that the dual implementation carries technical risk but argued that shipping both features together avoids the need for a separate hard fork and allows the two systems to be optimised as an integrated unit.
Additional Proposals Under Consideration
Beyond the two headline EIPs, more than 25 additional proposals remain under consideration for inclusion in Glamsterdam. Among the most closely watched is FOCIL, or fork-choice enforced inclusion lists, which would require block builders to include a minimum set of pending transactions in each block. FOCIL is designed to address censorship resistance concerns by preventing builders from systematically excluding certain transactions for economic or political reasons.
However, the Base engineering team at Coinbase has publicly warned that adding FOCIL alongside ePBS could delay Glamsterdam beyond 2026. The concern centres on the interaction between inclusion lists and the sealed-block mechanism in EIP-7732, which creates potential conflicts around transaction ordering and timing. Ethereum core developer Tim Beiko has indicated that FOCIL may be deferred to the subsequent upgrade, tentatively named Hegota and slated for late 2026 or early 2027.
Market and Ecosystem Implications
Glamsterdam arrives at a pivotal moment for Ethereum. The network's share of total DeFi value locked has declined from 58% to 52% over the past year as Solana, Base, and Arbitrum have captured an increasing share of developer and user activity. The upgrade's throughput improvements and fee reductions are widely viewed as essential for maintaining Ethereum's competitive position, particularly as Layer 2 networks continue to migrate high-frequency trading and consumer-facing applications off the mainnet.
The ePBS component also carries significant implications for the MEV supply chain, a market that generated an estimated $1.2 billion in extracted value during 2025. By eliminating reliance on third-party relays, Glamsterdam could redistribute MEV revenue more equitably between builders and proposers while reducing the systemic risk posed by relay centralisation. Flashbots has publicly supported the transition, stating that enshrined PBS aligns with its long-term vision for a decentralised and transparent block production pipeline.
Timeline and Risks
The on-time delivery of Ethereum's two most recent upgrades — Pectra in March 2025 and Fusaka in January 2026 — has bolstered confidence in the Foundation's improved execution cadence. However, Glamsterdam's scope exceeds both predecessors, and the interplay between ePBS and BALs introduces novel failure modes that will require extensive testing. Public testnets are expected to begin in April and May, with dual audit phases running through early June. If the audits identify critical issues, the June target could realistically slip to the third or fourth quarter. Developers have emphasised that shipping a secure upgrade takes precedence over meeting any specific date.
