Ethereum's next hard fork, Glamsterdam, is in Devnet-5 testing with an aspirational June 2026 mainnet target. The upgrade bundles two headliner EIPs — enshrined Proposer-Builder Separation (EIP-7732) and Block-Level Access Lists (EIP-7928) — alongside a gas repricing package (EIP-7904/EIP-8007) p...
"2026 [should] be the year Ethereum takes back lost ground in terms of self-sovereignty and trustlessness." — Vitalik Buterin, Co-Founder, Ethereum
Ethereum's next hard fork, Glamsterdam, is in Devnet-5 testing with an aspirational June 2026 mainnet target. The upgrade bundles two headliner EIPs — enshrined Proposer-Builder Separation (EIP-7732) and Block-Level Access Lists (EIP-7928) — alongside a gas repricing package (EIP-7904/EIP-8007) projected to cut fees by 78.6% and raise the block gas limit from 60 million to 200 million. If delivered on schedule, Ethereum L1 throughput would approach 10,000 transactions per second, a 10x increase from current capacity.
The fork carries significant economic implications. With ~1.1 million active validators staking 35.9 million ETH (28.9% of supply) at yields of 2–4% APY, and L2 sequencers collectively earning ~$161 million per year — revenue that largely bypasses L1 validators — Glamsterdam's ePBS mechanism could restructure the value chain by internalizing block-building economics into the protocol itself. The upgrade arrives as Ethereum trades near $2,040, with L1 fee revenue under pressure from its own scaling success: Layer 2 networks now process over 1.9 million daily transactions, pushing mainnet fees below $0.01 per transaction.
Glamsterdam is Ethereum's first hard fork since Fusaka and the largest structural upgrade since the Merge. Vitalik Buterin outlined eight EIPs in late February 2026 that define its scope, targeting three areas: block construction decentralization, execution-layer throughput, and gas cost alignment with actual computational overhead.
The upgrade is split across two layers. On the consensus layer, EIP-7732 introduces enshrined Proposer-Builder Separation. On the execution layer, EIP-7928 implements Block-Level Access Lists to enable parallel transaction processing. A bundle of gas repricing EIPs — anchored by EIP-7904 and aggregated in EIP-8007 — realigns opcode costs with benchmarked hardware workload profiles.
Additional proposals under consideration include EIP-7932, which introduces an upgrade path to quantum-resistant cryptographic schemes, and EIP-7980, which adds Ed25519 signature support for cross-chain interoperability. Over 25 non-headliner EIPs remain under review for inclusion, according to Ethereum Foundation documentation.
Today, Ethereum's block production relies on an off-chain relay ecosystem. Block builders construct blocks, submit them through relays like Flashbots, and validators (proposers) select the highest-bid block without seeing its contents. This system works but depends on a small number of relay operators — a centralization point that Ethereum developers have long flagged.
ePBS moves this market into the protocol itself. Under the new mechanism, builders cryptographically seal their blocks and commit to a bid on-chain. Validators select the highest bid through protocol rules rather than through off-chain relay infrastructure. The block contents are revealed only after the commitment is locked.
The economic implications are direct. MEV-extracting entities generated approximately $24 million in profit over a 30-day period from December 2025 to January 2026 on Ethereum alone. Under the current relay-dependent system, a significant portion of MEV value accrues to off-chain intermediaries. ePBS eliminates the relay dependency and opens block-building competition to a wider set of participants operating within protocol-enforced rules.
The censorship resistance argument is equally concrete. With fewer off-chain chokepoints, filtering transactions becomes structurally more difficult. According to the Ethereum Foundation's Platform team blog post from March 23, 2026, L1's role is to serve as a "truly permissionless and maximally resilient global hub for settlement, shared state, liquidity, and DeFi."
BALs are structured lists pre-declaring which storage slots and accounts each transaction in a block will access. This allows Ethereum nodes to identify non-conflicting transactions and process them simultaneously — parallel execution at the block level.
Current Ethereum execution is sequential: each transaction runs one after the next. BALs change this by providing the execution client with a map of state access before processing begins. Non-overlapping transactions run concurrently.
Combined with the gas limit increase from 60 million to 200 million, BALs are the primary mechanism behind the 10,000 TPS throughput target. The gas limit increase will be phased: 100 million initially, scaling to 200 million once ePBS is fully operational.
Since Ethereum's inception, computational gas costs have remained largely static even as EVM implementations have been extensively optimized. EIP-7904 recalibrates opcode gas costs using benchmarks from multiple production clients, anchoring prices to actual hardware workload profiles rather than theoretical estimates from 2015.
EIP-8007 aggregates the full Glamsterdam gas repricing package. The combined effect: a projected 78.6% reduction in gas costs for both simple ETH transfers and complex smart contract interactions. At current fee levels — where average mainnet transaction costs already hover near $0.01–$0.41 depending on complexity — this pushes L1 fees toward near-zero for standard operations.
The repricing addresses a long-standing misalignment. Expensive compute operations like hashing and precompiles were underpriced relative to their actual resource consumption, while simpler operations were overpriced. The recalibration aims to prevent state bloat by making storage-heavy operations appropriately costly while reducing friction for computation-light transactions.
As of early April 2026, Glamsterdam is in Devnet-5 testing. The Ethereum Foundation DevOps team completed testing of three proposed EIPs on Devnet-4 before transitioning to the current phase. Public testnet deployments with dual audit phases are planned for spring 2026.
The most recent All Core Devs Execution (ACDE) call — ACDE #233 on March 26, 2026 — covered devnet updates, benchmarking progress, EIP-8070 status, and initial discussion of Hegotá headliner selection, indicating that developers are already looking past Glamsterdam.
Core developers have stated publicly that hitting the June target date is secondary to correctness. A slip into Q3 2026 remains plausible. The scope reduction that moved FOCIL to the subsequent upgrade (Hegotá) was specifically motivated by timeline concerns.
Glamsterdam lands in the middle of an unresolved economic tension. Ethereum's scaling success through Layer 2 networks has come at a cost to L1 revenue.
The numbers tell the story:
Ethereum mainnet still generates roughly 2x the fee revenue of the second-place chain (Tron), but the trajectory is clear: cheap data availability — essential for L2 viability — undermines L1 fee revenue.
The Ethereum Foundation's March 2026 blog post attempts to reframe this dynamic. L2s should differentiate on "features, services, customizations, go-to-market strategies, and zones of control," while L1 serves as the settlement and liquidity hub. The EF formed a new Platform team in February 2026 specifically to manage L1-L2 coordination.
If Glamsterdam's 10x throughput increase makes L1 fast enough for many use cases currently handled by L2s, the competitive dynamic shifts. As one analysis noted: "pure-scaling L2s without differentiation will not survive." Only 2 of more than 50 major L2s had reached Stage 2 decentralization by early 2026, per Buterin's own assessment.
EIP-7918, a blob fee floor mechanism, represents a partial fix to the L1 revenue bleed, but the fundamental tension between cheap data availability and L1 economic sustainability remains unresolved.
Fork-Choice Inclusion Lists (FOCIL) — a censorship resistance mechanism — was originally considered for Glamsterdam but was moved to Hegotá, the subsequent upgrade expected in H2 2026.
The Base engineering team publicly warned that bundling FOCIL with ePBS could delay Glamsterdam beyond 2026 entirely. Developers agreed to split the scope: ePBS and BALs ship in Glamsterdam, FOCIL ships as Hegotá's sole headliner feature.
This creates a gap. ePBS addresses centralization in block construction, but without FOCIL, there is no protocol-level mechanism forcing proposers to include specific transactions. The censorship resistance improvement from ePBS alone is indirect — more builders can compete, making filtering harder — but not absolute. Full censorship resistance requires both mechanisms working in concert.
Hegotá is tentatively scheduled for H2 2026, meaning Ethereum could have both ePBS and FOCIL operational before year-end. Whether this timeline holds depends on Glamsterdam shipping without significant delay.
Approximately 1.1 million active validators currently secure Ethereum, with 35.9 million ETH staked (28.9% of total supply). Average staking yields range from 2–4% APY depending on method and MEV-Boost usage.
Glamsterdam affects validators through several channels:
ePBS restructures MEV flow. By eliminating off-chain relay dependency, ePBS changes how MEV value is distributed. Validators interact with a protocol-native builder market rather than relying on Flashbots or similar intermediaries. The net effect on validator revenue depends on how competition evolves within the enshrined system.
Gas limit increase raises hardware requirements. A 200 million gas limit requires nodes to process significantly more data per block. The BALs mechanism partially offsets this by enabling parallel execution, but validators running minimal hardware may face pressure to upgrade. CryptoSlate flagged this as a risk in a recent analysis of the 2026 roadmap.
Fee reduction lowers priority fee income. With a 78.6% reduction in gas costs, the absolute value of priority fees (tips) paid to validators decreases proportionally, all else being equal. If throughput increases compensate through higher transaction volume, the net effect could be neutral or positive. Data is inconclusive at this stage.
Glamsterdam represents Ethereum's attempt to reclaim L1 relevance in a network increasingly defined by its Layer 2 ecosystem. The technical changes are substantial: protocol-native block building, parallel execution, and the first comprehensive gas repricing since genesis. Whether the 10,000 TPS target materializes depends on successful testing through spring and summer 2026.
The economic stakes are high. With L2 payments to L1 down over 90% year-over-year and sequencer revenue concentrating in a handful of rollups, Ethereum L1 must demonstrate that scaling the base layer creates more value than it cannibalizes from its own extension networks. Glamsterdam's ePBS mechanism is a structural response to off-chain centralization, but its revenue impact on validators remains uncertain until the enshrined builder market establishes competitive equilibrium.
The next milestones: public testnet deployments in spring 2026, dual audits, and a mainnet activation window that developers have deliberately left flexible.