Ethereum's Glamsterdam hard fork — a bundle of ten Ethereum Improvement Proposals tracked under Meta EIP-7773 — entered its final devnet phase in mid-June 2026 and targets mainnet activation in H2 2026. The upgrade's two headline proposals, EIP-7732 (Enshrined Proposer-Builder Separation) and EIP...
"This is probably the largest fork we've had since the Merge." — Parithosh Jayanthi, Ethereum Foundation DevOps Engineer
Ethereum's Glamsterdam hard fork — a bundle of ten Ethereum Improvement Proposals tracked under Meta EIP-7773 — entered its final devnet phase in mid-June 2026 and targets mainnet activation in H2 2026. The upgrade's two headline proposals, EIP-7732 (Enshrined Proposer-Builder Separation) and EIP-7928 (Block-Level Access Lists), restructure how blocks are built and executed at the protocol level. Together they lay the foundation for a gas-limit increase from the current ~60 million to a target of 200 million per block — roughly tripling Layer 1 throughput — and a theoretical ceiling near 10,000 transactions per second.
The fork arrives at a precarious moment for Ethereum. ETH trades near $1,570, down from a $2,100 band earlier in H1 2026. DeFi TVL on Ethereum has contracted 43% year-to-date to $38.9 billion, according to DefiLlama. L1 daily fee revenue has fallen to single-digit millions, a fraction of the ~$30 million generated during the 2021-2022 cycle. Glamsterdam is, in concrete engineering terms, Ethereum's attempt to rebuild Layer 1 economic throughput without abandoning the rollup-centric roadmap.
Glamsterdam is the direct successor to the Fusaka fork (December 2025) and the largest coordinated protocol change since the Merge in September 2022, according to Ethereum Foundation engineers. The upgrade ships ten EIPs organized across four categories:
Scale L1 and Parallel Processing:
Network Sustainability:
Network Resilience:
User and Developer Experience:
Additional proposals under testing include EIP-7778, EIP-7843, EIP-7976, EIP-7981, and EIP-8024. Core developers declined to include 6-second slots and FOCIL inclusion lists for this upgrade due to integration complexity, deferring both to the next fork, Hegotá.
EIP-7732 targets a structural dependency that has grown into an unacknowledged systemic risk. Approximately 88-93% of Ethereum blocks are currently built off-chain through MEV-Boost, according to data compiled by mevboost.pics and referenced in ethereum.org documentation. Five entities relay 98% of MEV-boosted transactions. Following Blocknative's exit in 2023, four major relayers remain active, and four builders construct nearly 50% of all blocks on the network.
This concentration means that most of Ethereum's block production depends on off-protocol infrastructure maintained by a small number of private operators. MEV-Boost is not enshrined in Ethereum's consensus rules; it is middleware that validators voluntarily adopt to access higher-value blocks from specialized builders. The arrangement works, but it creates single points of failure, censorship vectors, and opaque value extraction.
ePBS moves the proposer-builder separation mechanism directly into the consensus layer. Under the new design:
The payload propagation window expands from approximately 2 seconds to roughly 9 seconds. This is not merely a timing change. The wider window means blocks can contain more data — both transactions and blobs — without risking propagation failures that would orphan the block. MEV-Boost relays become optional rather than required, and builder payments settle natively on-chain rather than through trusted intermediaries.
EIP-7928 introduces Block-Level Access Lists (BALs), which map every account and storage slot touched by each transaction in a block. This metadata enables two capabilities:
Together, ePBS and BALs provide the technical foundation for the gas-limit increase to 200 million. The current ~60 million limit would be unsafe at higher thresholds under serial execution; parallel processing and the wider propagation window remove those constraints.
Glamsterdam includes three repricing EIPs that adjust gas costs to reflect actual resource consumption:
Counterbalancing these increases, EIP-2780 reduces intrinsic transaction gas costs. The net effect for users: standard ETH transfers between existing accounts could become up to 71% cheaper, according to Datawallet's analysis. The repricing steers usage toward blobs for data and away from calldata, aligning economic incentives with the network's architectural direction.
| Milestone | Date/Status | |-----------|-------------| | Soldøgn interop devnet concluded | May 2, 2026 | | Final devnet stage entered | Mid-June 2026 | | Public testnets (Holesky, Hoodi) | July-August 2026 (expected) | | Mainnet activation window | September-December 2026 |
Ethereum Foundation engineer Parithosh Jayanthi described the current phase as "the last step before hardening and shipping to public testnets." Core developers have emphasized that "getting ePBS right outranks any fixed date," according to Datawallet reporting. Some staking providers cite an internal optimistic target around Q3 2026, but the firmer base case is September to December, given the two to four months of public-testnet seasoning recent forks have required.
Glamsterdam does not directly increase blob counts — that was primarily Fusaka's contribution via PeerDAS. However, the wider ePBS propagation window (9 seconds vs. 2 seconds) creates room for future blob expansion. More blobs per block means lower data costs for L2 networks that post transaction data to Ethereum for settlement guarantees.
This matters because the L2 ecosystem has grown substantially. Arbitrum, Optimism, Base, and zkSync collectively process millions of transactions daily, each posting compressed batch data to Ethereum mainnet. The cost of that data posting directly affects L2 transaction fees for end users.
If the gas limit reaches 200 million and blob capacity expands in subsequent forks, Ethereum's data-availability budget — the total amount of rollup data the network can absorb per block — increases proportionally. This is the mechanism through which L1 scaling translates into L2 cost reduction.
Ethereum's economic position has deteriorated through H1 2026:
The fee-revenue decline reflects a structural shift: value-generating activity has migrated to L2s, where fees accrue to sequencer operators rather than to ETH holders through base-layer burn. Glamsterdam's gas-limit tripling is an explicit attempt to pull transaction volume back to L1 or, at minimum, ensure that L1 captures more economic activity per block.
Two parallel institutional developments underscore the stakes. On July 1, 2026, a new non-profit called Ethereum Institutional launched with backing from BitMine Immersion Technologies (NYSE: BMNR), SharpLink (NASDAQ: SBET), and Ethereum co-founder Joe Lubin. The organization consolidates the Ethereum Foundation's prior go-to-market team into an independent entity focused on institutional engagement for tokenization, stablecoins, and on-chain market infrastructure. Ethereum currently hosts approximately $180 billion in stablecoins on mainnet — about 60% of total stablecoin supply — and roughly two-thirds of tokenized real-world assets, per the organization's launch statement.
Separately, Ethlabs launched in the same week as a non-profit for research and development. The two entities — Ethereum Institutional for commercial engagement and Ethlabs for technical development — represent a deliberate organizational split of functions previously housed within the Ethereum Foundation.
All consensus-layer (CL) and execution-layer (EL) clients must update before mainnet activation:
The new Payload Timeliness Committee duty requires CL client support. Validators who fail to update will be unable to participate in block proposal or attestation after the fork activates. Given the 39 million ETH staked across the network, the coordination requirement is substantial — late updates by major staking providers could cause attestation disruptions during the transition.
Glamsterdam is an infrastructure upgrade with direct economic implications. By enshrining proposer-builder separation and enabling parallel execution, it removes two technical bottlenecks that have constrained L1 throughput since the Merge. The 200-million gas-limit target, if achieved, would represent the most significant capacity expansion in Ethereum's history.
Whether this translates into fee-revenue recovery depends on demand. A tripled gas limit means nothing if transaction volume does not follow. The concurrent launch of Ethereum Institutional signals that the ecosystem's leadership recognizes the problem is not purely technical — it is also commercial. Ethereum needs institutional transaction volume on L1 to justify its infrastructure investment, and it needs protocol-level changes to handle that volume safely.
The testnet phase over the coming months will determine whether the September-December mainnet window holds. Core developers have indicated they will not rush. For validators, client operators, and staking providers, the preparation window has begun.