Ethereum's Glamsterdam hard fork — tracked under Meta EIP-7773 — entered its final developer network phase in late June 2026, bundling ten Ethereum Improvement Proposals into what core developers call the most significant protocol change since the September 2022 Merge. The upgrade targets mainnet...
"We're working on devnets with all the EIPs in them right now. This is the last phase before we work on hardening and then shipping the testnets." — Parithosh Jayanthi, DevOps Engineer, Ethereum Foundation
Ethereum's Glamsterdam hard fork — tracked under Meta EIP-7773 — entered its final developer network phase in late June 2026, bundling ten Ethereum Improvement Proposals into what core developers call the most significant protocol change since the September 2022 Merge. The upgrade targets mainnet activation between September and December 2026, pending two to four months of public testnet validation.
Two headline proposals define the fork. EIP-7732 enshrines Proposer-Builder Separation (ePBS) directly into the consensus layer, replacing the off-protocol MEV-Boost relay system that currently routes roughly 90% of Ethereum blocks through three dominant builders controlling 84% of production. EIP-7928 introduces Block-Level Access Lists (BALs), enabling parallel transaction execution by mapping every account and storage slot a block touches before processing begins. Together with a gas-repricing package under EIP-7904, the bundle clears the path for a validator-coordinated increase from the current 60 million gas limit to a 200 million gas-limit floor — a roughly 3.3x expansion of L1 throughput capacity.
The economic implications are direct. Gas repricing alone projects a 78.6% reduction in execution costs for both simple transfers and complex smart contract interactions. A Uniswap trade currently costing $3–8 could drop below $1. Daily gas revenue, already compressed from a peak of $23 million to $6.3 million, will face further pressure. Whether cheaper block space generates sufficient volume to offset the per-unit fee decline remains the central question for ETH's value accrual thesis.
EIP-7773, the Glamsterdam meta proposal, lists ten EIPs as scheduled for inclusion. The bundle remains in draft status, meaning scope adjustments are still possible, though core developers indicated in June 2026 that the current set is what they expect to carry through testnets.
The full manifest:
| EIP | Function | |------|----------| | EIP-7732 | Enshrined Proposer-Builder Separation (ePBS) | | EIP-7928 | Block-Level Access Lists (BALs) | | EIP-7904 | Gas repricing based on current hardware benchmarks | | EIP-7708 | ETH transfers and burns emit a log | | EIP-7778 | Block gas accounting without refunds | | EIP-7843 | SLOTNUM opcode | | EIP-7954 | Maximum contract size raised from ~24 KiB to 32 KiB | | EIP-8024 | Backward-compatible SWAPN, DUPN, EXCHANGE opcodes | | EIP-8037 | State-creation gas-cost restructuring | | EIP-8159 | eth/71 Block Access List Exchange |
The two structural changes — ePBS and BALs — do the heavy lifting. The remaining eight proposals address operational improvements, gas accounting adjustments, and developer tooling enhancements.
Ethereum's block production today operates through a two-tier system. Validators propose blocks, but the actual construction of those blocks has migrated off-protocol to specialized builders operating through MEV-Boost relays maintained by Flashbots and a small number of other operators. As of July 2026, Titan produces 51.34% of MEV-Boost blocks, Quasar handles 18.37%, and Eureka accounts for 14.69%. Five builders collectively control 96.7% of all MEV-facilitated blocks.
This concentration creates three identified risks: censorship capability (only three of seven major relays do not filter transactions per OFAC compliance), single-point-of-failure fragility if a dominant builder exits, and a trust layer between proposers and builders that the protocol does not govern.
EIP-7732 moves the proposer-builder separation mechanism into the consensus layer itself, eliminating the need for trusted relay intermediaries. Under ePBS, any entity meeting staking requirements can participate as a builder. The mechanism creates a permissionless market for block construction within the protocol.
The limitation is structural. Academic research from papers published in May and June 2026 (arXiv) indicates that while ePBS decentralizes at the proposer level, it may intensify builder concentration. Capital-intensive builders with sophisticated MEV extraction infrastructure retain competitive advantages. Exclusive orderflow arrangements — where transaction senders route directly to preferred builders — persist under ePBS unless separately addressed. Vitalik Buterin has acknowledged this, publicly exploring "big FOCIL" (Fork-Choice enforced Inclusion Lists) and encrypted mempools as complementary mechanisms to prevent centralization in the block-building pipeline.
EIP-7928 introduces Block-Level Access Lists (BALs): a data structure included in every block that maps every account and storage slot the block will touch, along with post-execution state values. This serves as a pre-execution blueprint.
The practical effect: because BALs reveal which transactions operate on non-overlapping state, nodes can safely perform parallel disk reads and group independent transactions for simultaneous processing. Ethereum's execution model has been strictly sequential since launch — every transaction processed one after another, regardless of whether they interact with the same state.
BALs build on a foundation laid by the Fusaka upgrade (December 2025), which shipped PeerDAS (Peer Data Availability Sampling) to scale rollup blob throughput. Where Fusaka addressed data availability for Layer 2s, Glamsterdam targets the execution engine for Layer 1.
The parallelism enabled by BALs is a prerequisite for safely operating at the 200 million gas-limit target. Without parallel execution, tripling block capacity would proportionally increase block processing time, degrading propagation and risking chain splits.
EIP-7904 recalibrates opcode gas costs to reflect actual computational costs on modern hardware. Many current gas prices were set based on benchmarks from 2016–2020 and no longer correspond to real execution costs. Some operations are significantly cheaper to execute than their gas price implies; others are more expensive and have served as attack vectors.
The recalibration produces a projected 78.6% reduction in gas costs across both simple ETH transfers and complex smart contract interactions. At current network conditions — with standard gas around 0.15 gwei and daily averages near 0.5 gwei through Q2 2026 — a basic ETH transfer already costs under a cent. Post-Glamsterdam, multi-step DeFi operations involving multiple contract calls would see proportionally larger savings.
The 200 million gas-limit floor is a design target, not a value the fork enforces. Validators set the gas limit through standard gas-vote signaling, currently coordinated around 60 million. The increase would proceed incrementally as node operators demonstrate they can handle larger blocks without degraded propagation. This mirrors the community-driven gas-limit increase from 30 million to 60 million that occurred progressively through 2024–2025.
EIP-7778 removes gas refund mechanics from block accounting, simplifying the execution model and eliminating a class of economic attacks where contracts manipulate refunds to effectively execute below-cost transactions.
Glamsterdam addresses a tension inherent in scaling: increasing compute capacity without proportionally expanding Ethereum's permanent state database. Two proposals target this directly.
EIP-8037, finalized in May 2026, sets a fixed cost per byte of new state and carves out a separate gas reservoir for state growth, decoupling state creation from execution gas. Until now, deploying a new contract or opening a new storage slot consumed the same gas type as running computations, meaning scaling compute capacity inevitably inflated state size.
EIP-7954 raises the maximum contract size from approximately 24 KiB to 32 KiB, a 33% increase that accommodates larger on-chain applications without requiring developers to split logic across multiple contracts — a practice that adds complexity, gas overhead, and attack surface.
The state-growth controls reflect a lesson from Ethereum's history: the 2016 Shanghai DOS attacks exploited underpriced state-creation operations. EIP-8037 aims to price state growth at its true long-term cost to the network — the cost of every full node storing that data indefinitely.
Ethereum's L1 fee economy is already under pressure. Daily gas revenue declined from a peak of approximately $23 million to $6.3 million through H1 2026. Over 4.5 million ETH have been burned through the EIP-1559 mechanism since August 2021, but the burn rate has slowed substantially as average gas prices compressed to sub-1 gwei levels.
Glamsterdam's gas repricing will reduce per-transaction fees further. The economic bet is volumetric: cheaper block space stimulates more on-chain activity, and the aggregate fee burn from higher transaction count offsets the per-unit decline.
Supporting this thesis: Ethereum's Q1 2026 data showed transaction volume reaching all-time highs even as gas fees declined. The network is adding users — 38.9 million ETH is staked (31.98% of supply) across roughly 897,000 active validators, with consensus-layer yields at 2.78% APR and all-in yields (including MEV) around 3.3–4%.
Working against it: Layer 2 rollups — which already reduced L1 fee capture significantly through blob-based data posting — will gain further from cheaper L1 settlement costs, potentially reducing their incentive to batch transactions and further compressing L1 revenue. The existing report on Ethereum's L2 value-capture dynamics documented a $100M annual fee leak from L1 to L2s.
The staking economics face a structural question. If ePBS reduces MEV extraction inefficiencies and distributes builder profits more broadly, the MEV component of validator yield (currently 0.5–1% supplemental APR) may compress. Conversely, higher throughput could expand the total MEV opportunity set.
The current development timeline, based on core developer statements and historical fork cadence:
Risk factors include:
Technical complexity. Glamsterdam ships more structural changes to the execution and consensus layers simultaneously than any fork since the Merge. ePBS and BALs interact in ways that devnet testing may not fully stress-test until public testnets handle adversarial conditions.
Builder migration. Transitioning from off-protocol MEV-Boost to enshrined PBS requires builder infrastructure to adapt. Incumbent builders may resist or exploit the transition period.
Gas-limit coordination. The 200 million target requires validator consensus through gas-vote signaling. A prolonged period of disagreement could result in an uneven or incomplete capacity expansion post-fork.
Scope creep. EIP-7773 remains in draft status. Historical precedent shows proposals being added or removed during the testnet phase — Pectra's scope was adjusted multiple times before its May 2025 launch.
Glamsterdam represents a structural overhaul of Ethereum's block production, execution, and fee economics. It addresses real, measurable problems: builder concentration (three entities controlling 84% of block production), sequential execution bottlenecks, and gas costs disconnected from actual computational expense.
The upgrade does not solve everything it touches. ePBS decentralizes proposer access but may not reduce builder concentration without complementary mechanisms. Parallel execution enables higher throughput but depends on validator coordination to actually raise gas limits. Gas repricing reduces costs but accelerates the fee compression already pressuring L1 revenue.
What it does accomplish is narrowing the gap between Ethereum's L1 capabilities and the demands being placed on it by a growing L2 ecosystem, institutional staking participants, and an expanding application base. Whether the resulting economics favor ETH holders depends on a volumetric bet that cheaper block space produces more total economic activity — a thesis that Q1 2026's all-time-high transaction volumes support but do not confirm at post-Glamsterdam price levels.
The data will arrive in H2 2026. Until then, the fork remains in testing.