Ethereum's next hard fork, Glamsterdam, has been pushed from its original June target to Q3 2026 — with end-of-August as the internal working date — following the Soldøgn interop in Svalbard on May 2. The upgrade raises the network's gas limit from 60 million to 200 million per block, representin...
"Ethereum has effectively been running in single-lane mode: one long queue of transactions, executed strictly in order. Even with multi-core CPUs, validation stays sequential. BALs change that." — Toni Wahrstätter, EIP-7928 Author & Ethereum Foundation Researcher
Ethereum's next hard fork, Glamsterdam, has been pushed from its original June target to Q3 2026 — with end-of-August as the internal working date — following the Soldøgn interop in Svalbard on May 2. The upgrade raises the network's gas limit from 60 million to 200 million per block, representing a 3.3x increase in execution capacity. If deployed on schedule, it would reduce gas fees by approximately 78% and push L1 throughput toward 10,000 transactions per second, up from the current ~20 TPS.
The upgrade is built around two headliner EIPs: EIP-7732 (Enshrined Proposer-Builder Separation) and EIP-7928 (Block-Level Access Lists). Together, they restructure how Ethereum builds and validates blocks, enabling parallel transaction processing for the first time in the network's history. This is Ethereum's most architecturally significant change since the Merge in September 2022.
The timing coincides with significant organizational flux at the Ethereum Foundation. Three new Protocol Cluster co-leads — Will Corcoran, Kev Wedderburn, and Fredrik — replaced outgoing leads Barnabé Monnot and Tim Beiko in May 2026, amid a broader exodus of at least eight senior staff members this year.
Vitalik Buterin outlined eight Ethereum Improvement Proposals for Glamsterdam in late February 2026. Two received top priority as "headliners": EIP-7732 on the consensus layer and EIP-7928 on the execution layer. Over 25 additional non-headliner EIPs remain under consideration, a scope that has raised concern among client teams.
The upgrade's name combines "Gloasjávri" (consensus layer, named after a lake in northern Norway) and "Amsterdam" (execution layer). Ethereum traditionally names its consensus forks after celestial bodies or geographic features and its execution forks after cities.
Core performance target: 60 million gas per second of execution throughput, per EIP-7904. At the post-Glamsterdam gas limit of 200M, this gives the network a 3x increase over its current ~20M gas/second baseline.
Currently, over 90% of Ethereum blocks are produced via MEV-Boost, a third-party relay system that separates block proposers from block builders outside the protocol. This works, but it introduces trust assumptions: proposers must rely on relay operators to honestly report bid values and deliver valid blocks.
EIP-7732 brings this separation into the protocol itself. Under ePBS:
The restructured slot timing expands the data propagation window from approximately 2 seconds to roughly 9 seconds. This expansion is critical: it gives the network headroom to safely propagate much larger blocks without increasing missed-slot rates.
MEV implications: Researchers estimate ePBS could reduce MEV extraction by up to 70%, according to Ethereum Foundation research. MEV searchers extracted approximately $24 million from Ethereum users in the 30-day window between late 2025 and early 2026. Removing relay middleware eliminates a centralization vector — currently, a small number of relays process the majority of block auctions.
Block-Level Access Lists (BALs) are arguably the more technically novel component. Today, Ethereum executes transactions strictly in sequence — each transaction must complete before the next begins, regardless of whether they touch the same state. Multi-core CPUs provide no benefit.
BALs attach a structured declaration to each block specifying exactly which accounts and storage slots every transaction will read or write. This enables:
Average overhead per block is approximately 35 KiB at 36M gas — below the worst-case sizes for calldata, making it bandwidth-feasible for the existing peer-to-peer network.
According to Ethereum research modeling, worst-case parallel execution under EIP-7928 still significantly outperforms sequential execution for typical mainnet block compositions.
EIP-7904 recalibrates gas costs to reflect actual computational resource consumption on modern hardware. Many existing gas prices were set years ago and no longer correspond to real execution costs.
Fee reduction estimates:
EIP-8037 complements the repricing by raising the cost of writing new state, preventing the higher gas limit from causing runaway state growth. The cost_per_state_byte parameter has been finalized following the Soldøgn interop.
The "ultrasound money" tension: A threefold capacity increase without matching demand growth would suppress base fees. Lower base fees reduce the EIP-1559 burn rate — the mechanism that has been central to ETH's deflationary narrative since the Merge. If L1 fees stay low, ETH issuance becomes net inflationary, compressing the monetary premium on the asset.
Counter-argument: a cheaper, faster L1 may pull activity back from Layer 2 networks and competitor chains, increasing total fee volume even as per-transaction costs drop.
Soldøgn Interop (May 2, 2026): Over 100 Ethereum core contributors gathered in Longyearbyen, Svalbard for focused coordination work. By the end of the week, the group achieved three core goals:
The multi-client devnet (glamsterdam-devnet-2) is running, with the external builders pipeline tested end-to-end.
Risk factors:
Internal target: End of August 2026 for mainnet activation.
The upgrade proceeds against a backdrop of significant organizational change. In May 2026, the Ethereum Foundation announced:
At least eight senior staff have left the foundation in 2026. Earlier in the year, co-executive director Tomasz Stańczak stepped down, with Bastian Aue named interim co-executive director alongside Hsiao-Wei Wang.
The leadership transition introduces execution risk for Glamsterdam. The new leads inherit an upgrade that is already in late-stage devnet testing, with limited margin for timeline slip before the community's patience erodes. ETH has dropped over 60% from its August 2025 all-time high of $4,951 to approximately $1,970 as of early June 2026.
ETH price context: Trading near $1,970 in early June 2026, ETH is under pressure from 17 consecutive days of ETF outflows and broader macro headwinds. Analyst year-end 2026 forecasts range from Citi's $3,175 to Standard Chartered's $7,500 — wide dispersion reflecting genuine uncertainty about whether Glamsterdam ships on time.
L2 impact: The 145 active Layer 2 protocols on Ethereum hold $17.9 billion in TVL as of Q1 2026. Post-Glamsterdam, L2s benefit from more blob space and lower settlement costs, but the cheaper L1 may also pull some activity back to mainnet — particularly for high-value DeFi operations where direct settlement provides superior security guarantees.
Competitive positioning: Solana processes 3,000–4,000 TPS today. A post-Glamsterdam Ethereum at even a fraction of its 10,000 TPS theoretical target would narrow the throughput gap significantly, though Ethereum's higher-value-per-transaction profile means raw TPS comparisons are incomplete.
Following upgrade (Hegotá): Will include FOCIL for censorship resistance, Verkle Trees to reduce node storage by up to 90%, and Account Abstraction improvements. No timeline set.
Glamsterdam represents Ethereum's attempt to resolve a longstanding architectural constraint — sequential transaction execution — while simultaneously enshrining the proposer-builder separation that currently operates outside the protocol via MEV-Boost. The combined effect is a threefold capacity expansion with parallel processing capabilities.
The technical foundations appear sound. Devnets are running, client teams have aligned on parameters, and FOCIL's deferral to Hegotá keeps scope manageable. The primary risks are organizational (leadership transition) and technical (ePBS coordination complexity at mainnet scale).
For the broader Ethereum ecosystem, the economic question is whether tripling capacity at 78% lower fees generates sufficient new demand to offset reduced per-transaction burn — or whether it accelerates the deflationary narrative's erosion. The answer will likely depend on whether a cheaper L1 can reclaim activity that migrated to competing networks during Ethereum's high-fee era.
End-of-August is the target. The market will price accordingly.