Ethereum's next hard fork, Glamsterdam, has cleared its final interoperability milestone. At a week-long developer gathering in Svalbard, Norway, core contributors locked in a post-upgrade gas limit floor of 200 million — more than five times the current 36 million ceiling — backed by three conve...
"The target was achieved through convergence of ePBS, BAL optimizations, and EIP-8037 repricing." — Ethereum Foundation, Protocol Cluster Updates: May 2026
Ethereum's next hard fork, Glamsterdam, has cleared its final interoperability milestone. At a week-long developer gathering in Svalbard, Norway, core contributors locked in a post-upgrade gas limit floor of 200 million — more than five times the current 36 million ceiling — backed by three converging technical workstreams: Enshrined Proposer-Builder Separation (ePBS), Block-Level Access Lists (EIP-7928), and state-cost repricing (EIP-8037).
The Ethereum Foundation confirmed on May 11 that ePBS has stabilized on a multi-client Glamsterdam devnet with the external builders pipeline tested end-to-end across nearly all execution and consensus clients. Mainnet activation remains officially targeted for H1 2026, though multiple developer sources now point to Q3 (July–September) as the realistic window. The upgrade would constitute Ethereum's most aggressive structural overhaul since The Merge in September 2022.
The economic implications are substantial. Two builders — Titan (52.16% of blocks) and BuilderNet (24.63%) — currently control nearly 80% of Ethereum's block production through the off-protocol MEV-Boost relay system. Glamsterdam's ePBS moves block auctions on-chain, reducing reliance on third-party middleware and potentially cutting MEV extraction by up to 70%, according to Ethereum researchers. For the 1 million+ validators securing the network, the upgrade restructures how execution-layer revenue is distributed.
Glamsterdam bundles three interlocking protocol changes. Each addresses a distinct bottleneck; together, they form a unified scaling and decentralization package.
| Component | EIP | Function | |---|---|---| | Enshrined Proposer-Builder Separation | ePBS (multiple EIPs) | Moves block building auctions into the consensus layer | | Block-Level Access Lists | EIP-7928 | Enables parallel transaction execution across CPU cores | | State Cost Repricing | EIP-8037 | Reprices storage operations with fixed cost_per_state_byte |
The upgrade also lays groundwork for Hegotá, Ethereum's planned late-2026 fork, which will introduce FOCIL (Fork Choice Inclusion Lists) to prevent validator-level transaction censorship and scope native account abstraction requirements.
The current block production pipeline depends on MEV-Boost, an off-protocol sidecar maintained by Flashbots. In this system, specialized builders construct execution payloads optimized for Maximal Extractable Value, and validators select the highest-paying block via relay intermediaries. Five relays currently handle the majority of Ethereum's block production: relay.ultrasound.money (33.92% of payloads), Titan relay (24.19%), bloXroute max-profit (14.67%), Aestus (10.03%), and bloXroute regulated (9.07%).
The builder market is even more concentrated. According to relay data as of April 2026, Titan controls 52.16% of all blocks built, BuilderNet handles 24.63%, and Quasar produces 15.06%. Three entities build approximately 92% of all Ethereum blocks.
ePBS eliminates the need for external relay infrastructure by enshrining the proposer-builder handoff directly into the consensus protocol. Under the new design, the proposer commits to a block header, and a separate builder constructs the execution payload within protocol-defined rules. The trustless exchange of a block payload for payment is handled natively, removing the dependency on MEV-Boost.
This architectural shift has two primary effects. First, it reduces the trust surface: validators no longer need to rely on relay operators to faithfully pass block headers and payments. Second, it introduces competition into the builder market at the protocol level, which Ethereum researchers estimate could reduce MEV extraction by up to 70%.
The implementation has proven technically demanding. According to the Ethereum Foundation's May 2026 protocol update, every part of the client stack must now reason about "partial blocks" and two-party coordination between proposers and builders — a change that touches practically every layer of the consensus implementation.
Ethereum currently processes transactions sequentially within each block. EIP-7928 introduces Block-Level Access Lists (BALs) — a new data structure appended to block headers via a block_access_list_hash field containing the Keccak-256 hash of the RLP-encoded access list.
BALs record all accounts and storage locations accessed during block execution, along with their post-execution values. This enables four categories of parallelization:
Historical chain data analysis shows that 60–80% of transactions in a typical block access disjoint storage slots, enabling effective parallelization without coordination overhead. The remaining 20–40% can still benefit from parallelization using post-transaction state diffs embedded in the BAL.
Average BAL size runs approximately 70 KiB per block, according to analysis of historical Ethereum data. The P2P protocol layer adds two new message types — GetBlockAccessLists and BlockAccessLists — enabling peers to request and serve BALs for synchronization.
Ethereum's gas limit has increased incrementally over five years:
| Period | Gas Limit | Mechanism | |---|---|---| | Pre-2021 | 15M | Validator consensus | | 2021 | 30M | Soft fork (validator signaling) | | Feb 2025 | 36M | Soft fork (>50% validator support) | | Mid-2025 | 60M | Soft fork (validator signaling) | | Post-Glamsterdam | 200M (target) | Hard fork + validator signaling |
The jump from 60M to 200M represents a 3.3x increase — the largest single leap in Ethereum's history. It is enabled by the confluence of parallel execution (EIP-7928 reducing per-transaction processing overhead), ePBS (optimizing block construction workflows), and EIP-8037 repricing (ensuring storage operations are priced correctly for the new throughput regime).
Full repricing numbers for EIP-8037 were delivered and validated on bal-devnet-6, the sixth iteration of Glamsterdam's dedicated testing network. The specification uses a fixed cost_per_state_byte model, replacing the previous ad-hoc gas schedule for storage operations.
Estimates from protocol researchers suggest that if the 200M gas limit is achieved alongside parallel execution, Ethereum Layer 1 could process approximately 10,000 transactions per second — roughly 10x current capacity. Gas fees could decline by approximately 79% at current demand levels.
According to the Ethereum Foundation's Protocol Cluster update published May 11, 2026:
The Foundation also announced a leadership transition for the Protocol Cluster. Barnabé Monnot, Tim Beiko, and Alex Stokes — the three coordinators who steered protocol development through Pectra, Fusaka, and the early Glamsterdam phase — are departing. Will Corcoran (research coordinator, zkVM proving and consensus), Kev Wedderburn (zkEVM team lead), and Fredrik Svantes (Protocol Security/Trillion Dollar Security lead) will assume coordination responsibilities.
Public testnet activations on Holesky and Sepolia are expected in the weeks ahead, pending devnet stability. The official target remains H1 2026, but developer commentary increasingly points to Q3 as the realistic activation window.
Glamsterdam restructures how economic value flows through Ethereum's block production pipeline.
Current state: Validators outsource block construction to builders via MEV-Boost relays. Builders extract MEV (sandwich attacks, arbitrage, liquidations) and share a portion with validators as a bid. Relay operators and searchers capture residual value. This off-protocol pipeline is opaque, concentrated, and introduces censorship vectors — some relays apply OFAC compliance filters to transactions.
Post-Glamsterdam: Builder-proposer coordination moves on-chain. The protocol enforces a trustless exchange: builders submit sealed bids, and proposers commit to headers without seeing execution payloads. Builder competition occurs within protocol rules rather than through external infrastructure.
For institutional stakers, the implications are mixed. The variable execution-layer component of staking rewards — currently driven by MEV-Boost tip payments — becomes subject to more competitive, protocol-mediated builder markets. More competition among builders generally means better block optimization over time, but individual MEV windfalls may decrease as extraction becomes harder. Staking rewards could become more predictable but potentially lower on a per-block basis.
The 79% fee reduction, if realized, would compress total fee revenue on Layer 1. Whether increased transaction volume from lower fees compensates for the per-transaction fee decline remains an open question. Ethereum's EIP-1559 burn mechanism means that higher throughput at lower fees could still maintain or increase the ETH burn rate — preserving the network's deflationary dynamics during periods of high activity.
ePBS complexity: The Ethereum Foundation's own assessment acknowledges that ePBS "touches practically everything" in the protocol stack. Reasoning about partial blocks and two-party coordination failure modes is architecturally novel. The history of complex protocol changes suggests extended testing periods are prudent.
Timeline risk: The H1 2026 target has already slipped in developer expectations to Q3. Each additional testnet issue or client incompatibility extends the timeline. The simultaneous departure of three Protocol Cluster coordinators during the most complex upgrade since The Merge introduces organizational risk.
Builder market response: Enshrining PBS does not guarantee decentralization of block building. If economies of scale in MEV extraction persist, the same dominant builders could control the on-chain market. The protocol change removes relay concentration risk but does not address builder-level concentration directly.
Gas limit governance: The 200M target is a floor, not a protocol-enforced parameter. Validators must signal support through client configuration. Whether sufficient validators will signal for 200M immediately post-fork, or whether the limit will increase gradually as it has historically, remains uncertain.
Glamsterdam is not an incremental upgrade. It rewires two foundational aspects of Ethereum simultaneously: how transactions are executed (parallel vs. sequential) and how blocks are constructed (on-chain auctions vs. off-protocol relays). The technical ambition is matched by the implementation complexity — ePBS alone touches nearly every layer of the client stack.
The economic stakes are significant. Approximately $2.4 billion in annual MEV flows through Ethereum's current block production pipeline, according to Flashbots data. Glamsterdam's ePBS does not eliminate MEV, but it changes who captures it and under what rules. For validators, stakers, and the broader DeFi ecosystem, the redistribution of this value — combined with dramatically higher throughput and lower fees — will reshape Ethereum's economic model.
Whether the upgrade ships in Q2 or Q3 2026 matters less than whether the multi-client testing validates the design under adversarial conditions. The departure of the protocol's most experienced coordinators during this critical phase adds an organizational variable that the Ethereum community will need to manage alongside the technical ones.