Ethereum's Glamsterdam hard fork — the network's largest protocol change since the 2022 Merge — has been officially pushed from its original June 2026 target to Q3 2026 following the Soldøgn interop devnet concluded May 2, 2026. The upgrade bundles eight Ethereum Improvement Proposals (EIPs) firs...
"In Glamsterdam, Ethereum is getting ePBS, which lets proposers outsource to a free permissionless market of block builders. This ensures that block builder centralization does not creep into staking centralization." — Vitalik Buterin, Ethereum Co-Founder, March 2, 2026
Ethereum's Glamsterdam hard fork — the network's largest protocol change since the 2022 Merge — has been officially pushed from its original June 2026 target to Q3 2026 following the Soldøgn interop devnet concluded May 2, 2026. The upgrade bundles eight Ethereum Improvement Proposals (EIPs) first outlined by Vitalik Buterin in late February 2026. If delivered as specified, Glamsterdam will raise the block gas limit from 60 million to 200 million, target throughput of approximately 10,000 transactions per second (roughly 10x current capacity), cut gas fees by an estimated 78.6% through computational repricing, and move block building on-chain via Enshrined Proposer-Builder Separation (ePBS).
The timing is significant. ETH trades at approximately $1,669 as of June 9, 2026 — down 60% from its August 2025 all-time high of $4,954. Spot Ethereum ETFs have recorded 17 consecutive days of net outflows, the longest streak for any crypto ETF product. May 2026 outflows totaled $401 million, the worst monthly reading since ETH ETFs launched. Eight senior Ethereum Foundation researchers departed in 2026, with five exits in May alone, raising questions about whether the organization retains the capacity to deliver the most complex upgrade in its history.
Glamsterdam is a two-layer upgrade. The consensus layer component, codenamed "Gloas," introduces Enshrined Proposer-Builder Separation. The execution layer component, codenamed "Amsterdam," delivers Block-Level Access Lists and gas repricing. Combined, they target three persistent criticisms of the Ethereum mainnet: gas fees are too high, throughput is too slow, and block building is too centralized.
The upgrade follows Pectra (May 2025) and Fusaka (late 2025), both of which shipped on time. Glamsterdam represents the third hard fork in roughly 14 months — a pace that the Ethereum Foundation has described as its target cadence of two major upgrades per year.
Current Ethereum Layer 1 handles approximately 1,000 TPS. The combined Layer 2 ecosystem processes roughly 5,600 TPS. Glamsterdam's 200-million gas limit target would bring L1 capacity to approximately 10,000 TPS under optimized conditions, though developers caution this is a theoretical upper bound rather than a guaranteed steady-state figure.
Buterin outlined the scope in a February 2026 post. The two headline proposals:
EIP-7732 (ePBS): Moves proposer-builder coordination from off-chain MEV-Boost relays into the Ethereum protocol itself. Block builders assemble and cryptographically seal blocks. Proposers select the highest-paying sealed block without visibility into its contents. Transactions are revealed only after finalization.
EIP-7928 (Block-Level Access Lists): Attaches to every block a deterministic record of which accounts and storage slots the block touches, along with post-execution state values. This enables clients to pre-fetch read/write sets, unlocking parallel execution, batched I/O, and parallel state-root computation.
Supporting EIPs include:
EIP-7904 (Gas Cost Repricing): Realigns gas costs to reflect actual computational resource consumption on modern hardware. Many current gas prices were set years ago and no longer correspond to execution costs. The result is an estimated 78.6% fee reduction across both simple ETH transfers and complex smart contract interactions.
Additional EIPs address quantum resistance improvements, enhanced user experience, and network-layer privacy features including FOCIL (a mechanism where 16 randomly selected attesters nominate transactions for mandatory inclusion) and encrypted mempool support.
The rationale for EIP-7732 is quantifiable. According to research published in late 2024 on arxiv (and corroborated by ESMA's 2025 paper on MEV), three dominant block builders construct approximately 80% of MEV-Boost blocks on Ethereum. Market share positively correlates with order flow diversity. Profitability correlates with access to exclusive order flow from integrated searchers. The result is a self-reinforcing oligopoly: builders need differentiated order flow to profit, but receive such flow only if they already hold significant market share.
This dynamic creates censorship risk. Concentrated block builders can selectively exclude transactions. ESMA flagged this as an operational and systemic risk. The current relay infrastructure — dominated by Flashbots — functions as middleware that the protocol cannot directly govern.
ePBS addresses this by separating the consensus role (proposing) from the execution role (building) at the protocol level. The data propagation window expands from 2 seconds to approximately 9 seconds, giving the network more time to validate and propagate blocks. This unblocks Ethereum's ability to handle higher transaction throughput while accommodating additional data blobs for Layer 2 networks.
The Flashbots team has separately proposed BuilderNet, a decentralized block building network running on hardware-secured enclaves, as a complementary approach. Whether ePBS alone is sufficient to break the builder oligopoly or whether supplementary infrastructure like BuilderNet is needed remains an open question.
EIP-7904 recalibrates gas costs based on empirical measurement of EVM operation costs. The Ethereum Foundation's analysis found that operations priced years ago do not reflect execution on modern hardware — some are significantly underpriced (creating bottlenecks), while others are overpriced (inflating user costs).
Key parameters post-Glamsterdam:
| Metric | Pre-Glamsterdam | Post-Glamsterdam | |---|---|---| | Block gas limit | ~60 million | 200 million | | Target L1 TPS | ~1,000 | ~10,000 (theoretical max) | | Gas fee reduction | Baseline | -78.6% (estimated) | | Data propagation window | ~2 seconds | ~9 seconds | | MEV block building | Off-chain (relays) | On-chain (ePBS) |
For Layer 2 economics, the implications compound. L2 transaction fees fell below $0.01 on major networks in early 2026 after Fusaka tripled blob capacity. EIP-7918 introduced a reserve price for blob fees to prevent fees from collapsing to near-zero (as occurred throughout 2024-2025, when blob base fees sat at or near 1 wei). Industry estimates suggest blob fees could contribute 30-50% of total ETH burn by the end of 2026, depending on L2 scaling. Core developers are planning for 48 blobs per block by mid-2026, with a long-term target of 128 blobs per slot under full Danksharding.
The Soldøgn interop devnet concluded May 2, 2026, producing a stable multi-client Glamsterdam devnet. Devnet-4 testing is complete. Devnet-5 is underway. Public testnets — Sepolia and Hoodi — are expected before mainnet activation.
The Ethereum Foundation confirmed Q3 2026 as the revised target after the Soldøgn interop, with testnet activations serving as the clearest leading indicators for the mainnet date. The established 200-million gas limit floor and stabilized ePBS implementation were described as key milestones achieved during Soldøgn.
The delay from June to Q3 is not necessarily a negative signal. Pectra and Fusaka both shipped without critical post-activation bugs. Developers have publicly stated that schedule fidelity is secondary to getting the upgrade right, a position consistent with past Ethereum upgrade management.
Overlaying Glamsterdam's technical ambition is an organizational question. At least eight senior Ethereum Foundation researchers departed in 2026. Five left in May alone. The departures include:
The Foundation's "Protocol Cluster" — the team responsible for Ethereum protocol research — has lost contributors across every layer it covers. Buterin has defended the transition, writing that "the EF is choosing to use its remaining resources to pursue longevity over breadth" and that the departures reflect a "deliberate generational transition, not institutional failure."
Critics counter that losing protocol coordinators and MEV researchers during the delivery window of an upgrade whose centerpiece is MEV reform (ePBS) creates execution risk. The Foundation has offered limited public explanation of its succession plan.
Notably, Beiko authored the Soldøgn interop post in May 2026, suggesting at least partial continuity through the testing phase.
ETH opened June 2026 near $1,975 and fell to $1,505 on June 6 — its weakest level since early 2023. The token has shed more than 60% from its August 2025 all-time high of $4,954.
Spot Ethereum ETF data:
Standard Chartered cut its 2026 ETH price target by 47% to $4,000, citing ETF outflows and macro headwinds. The outflow streak is longer than any comparable pattern recorded for Bitcoin ETFs, according to SoSoValue data, suggesting the institutional rotation is Ethereum-specific rather than a broad crypto-market move.
Contributing factors extend beyond the brain drain. Elevated interest rates, rotation from crypto into AI equities (with Anthropic and SpaceX IPOs drawing capital), and Ethereum co-founder Buterin selling ETH in early 2026 have all suppressed institutional demand.
The Ethereum Foundation aims to maintain its twice-a-year cadence with Hegotá scheduled for H2 2026. The upgrade's primary focus is Verkle Trees — a data structure designed to reduce node storage requirements by approximately 90%, enabling stateless clients. Hegotá also targets state expiry mechanisms to limit storage bloat.
If both Glamsterdam and Hegotá ship in 2026, Ethereum will have completed four hard forks in approximately 18 months. Whether the Foundation's reduced headcount can sustain this pace is an unanswered question.
Glamsterdam targets 10,000 TPS with a 200M gas limit and 78.6% fee reduction — the most significant execution layer overhaul since the Merge. Delivery is now Q3 2026 after being pushed from June.
ePBS (EIP-7732) addresses block builder centralization where three builders control approximately 80% of MEV-Boost blocks. The change moves block building on-chain and cryptographically separates building from proposing.
Block-Level Access Lists (EIP-7928) enable parallel execution — a structural shift from Ethereum's single-threaded 2015 architecture to deterministic, parallelizable state access.
Eight senior EF researchers departed in 2026, including the protocol coordinator and MEV researchers central to Glamsterdam's design. Organizational continuity during delivery remains an open risk.
ETH is trading 60% below its August 2025 all-time high with 17 consecutive days of ETF outflows totaling $401M in May 2026. The upgrade is the primary technical catalyst on the near-term horizon.
Hegotá (H2 2026) follows with Verkle Trees targeting 90% node storage reduction, but depends on a Foundation team that is demonstrably smaller than when these upgrades were designed.
Glamsterdam represents a technically sound response to Ethereum's most persistent criticisms: high fees, limited throughput, and centralized block production. The eight EIPs are well-specified, the devnet is stable, and the testing cadence suggests Q3 2026 delivery is plausible. The 10,000 TPS target and 78.6% fee reduction, if realized, would narrow the performance gap between Ethereum L1 and competing chains that have used throughput advantages as their primary marketing argument.
The risk is organizational, not technical. Ethereum's protocol research bench is thinner than at any point since the Merge. The Foundation lost its protocol coordinator, its MEV researchers, and several consensus-layer specialists in the same quarter it is expected to finalize the most complex upgrade since proof-of-stake. Buterin's framing — that this is a planned transition to a smaller, more focused organization — is internally consistent but untested. The market, reflected in a 60% drawdown and record ETF outflows, is pricing uncertainty, not conviction.
Whether Glamsterdam ships cleanly in Q3 2026 will serve as a definitive test of whether Ethereum's decentralized development model can sustain delivery without centralized organizational continuity.