Glamsterdam Reaches Its Final Stretch

Ethereum developers have entered the final development phase of Glamsterdam, the network's next major upgrade, with teams now testing a version of the fork that includes all the planned protocol changes. Developers are currently running developer networks, known as devnets, that contain the full suite of Ethereum Improvement Proposals slated for the upgrade. This marks what they describe as the final phase before the codebase is hardened and deployed to public testnets.

The progress was confirmed by Parithosh Jayanthi, a core developer and devops engineer at the Ethereum Foundation. He said the team is working on devnets with all the EIPs in them right now, calling it the last phase before hardening and shipping the testnets. While there is no fixed timeline, Jayanthi said the team has made massive progress.

Jayanthi described Glamsterdam as probably the largest fork since the Merge, the network's 2022 transition to proof-of-stake. He added that the upgrade will change a lot of assumptions about Ethereum and set the network up for much more scaling in the future. A firm activation date has yet to be determined, but Glamsterdam is currently expected to go live during the second half of 2026.

What Glamsterdam Changes

Glamsterdam represents a significant shift back toward Layer 1 structural improvements, following the successful rollout of the Fusaka upgrade in December 2025. The upgrade introduces structural changes to Ethereum's core architecture that will reshape the experience for everyone from everyday users and dApp developers to the validators and node operators who secure the network.

Two headline features anchor the upgrade. The first is enshrined Proposer-Builder Separation, known as ePBS. This formalizes the separation between the validators who propose blocks and the specialized builders who assemble them, baking the arrangement directly into the protocol. The Ethereum Foundation has reported that ePBS is now running stably on a multi-client Glamsterdam devnet, with the external block builder process having completed end-to-end testing across almost all client implementations. The change allows separate builder processes to assemble blocks while proposers focus on consensus, a key step in formalizing the MEV supply chain at the protocol level.

The second headline feature is Block-Level Access Lists, formally tracked as EIP-7928. BALs allow the network to pre-fetch data and execute transactions in parallel, increasing efficiency across the base layer. By enabling parallel execution and potentially raising the gas limit, the upgrade aims to scale Layer 1 throughput substantially while improving decentralization and fairness around miner extractable value.

The Technical Details Behind the Upgrade

The parallel execution that BALs enable is one of the most consequential changes in Glamsterdam. By allowing the network to pre-fetch data and process transactions simultaneously rather than strictly in sequence, the upgrade targets a meaningful increase in Layer 1 efficiency. Analysis of the upgrade has projected that increasing efficiency and raising the gas limit toward 200 million could reduce Layer 1 gas costs significantly, making the base layer more affordable during high-traffic periods.

Another component, EIP-8037, increases the gas cost of state creation and has reached its final draft, parameterized in one of the test networks. The proposal introduces a fixed cost per state byte model designed to target state growth at a controlled rate, with contract deployment costs rising roughly tenfold and new account creation about 8.5 times, while maintaining separate metering for code deposit so large contracts like Uniswap pools remain deployable. This addresses the long-term challenge of state bloat, where the ever-growing size of Ethereum's state data places increasing burdens on node operators.

The Glamsterdam candidate list initially exceeded 50 EIPs, but the final cut reflects deliberate scope discipline by Ethereum core developers. Some features originally considered for Glamsterdam, including certain scalability and censorship-resistance components, have been moved to Hegotá, Ethereum's subsequent major upgrade. According to the Foundation's summary, the combined results of the ePBS work, BAL optimization, and EIP-8037 repricing have given client teams a trusted path for Glamsterdam's final scope.

The Path to Mainnet

The route to mainnet activation runs through a structured sequence. Developers run iterative devnets to trial new code, then move to stabilization, security review, and public testnet readiness before any mainnet fork date is announced. The Soldøgn interop devnet concluded on May 2, 2026, with a stable multi-client Glamsterdam devnet now running. Public testnets, expected to include networks like Sepolia and Hoodi or their equivalents, are expected to follow before mainnet activation.

While a firm activation date has not been locked in, internal working targets have pointed toward the end of August 2026, though developers have consistently stressed that this is aspirational and dependent on devnet readiness and cross-client testing. The Ethereum Foundation has been explicit that hitting any particular date is secondary to getting the upgrade right.

Several risks could push activation later into 2026. The complexity of enshrining proposer-builder separation has proven trickier than anticipated and remains the primary bottleneck to testnet readiness. All major execution layer and consensus layer clients must also reach feature parity before a mainnet date can be confirmed. And while the gas repricing parameters were finalized, testing the repricing under realistic mainnet-scale load remains an outstanding workstream.

The Naming Convention and Coordinated Upgrade

The name Glamsterdam reflects Ethereum's tradition of combining a consensus layer star name with an execution layer city name. Amsterdam, which hosted Devconnect in 2022, serves as the execution layer anchor for this cycle. The combined name signals a coordinated upgrade across both layers, where the consensus layer and execution layer must activate together. That coordination is why the naming convention matters to client teams and node operators, who need both halves of the network to move in lockstep.

Ethereum has moved to a predictable, biannual upgrade schedule, with Glamsterdam succeeding Fusaka and preceding Hegotá. Glamsterdam and Hegotá are designed as complementary halves of the same vision. Glamsterdam handles execution-side improvements like faster processing, cheaper gas, and decentralized block building, while Hegotá addresses storage and state management. Together, they form one of Ethereum's most comprehensive periods of infrastructure development since the transition from proof of work.

Why It Matters for Users and the Ecosystem

For everyday Ethereum users, no manual action is required to benefit from Glamsterdam, but the invisible benefits are substantial, particularly during high-traffic events. Lower Layer 1 gas costs and higher throughput would make the base layer more usable and affordable, easing the congestion that has historically pushed activity onto Layer 2 networks and alternative chains.

The MEV-related changes from ePBS also carry significance for the network's fairness and decentralization. By formalizing the separation between proposers and builders at the protocol level, Ethereum aims to reduce the centralizing pressures that the MEV supply chain has introduced, where sophisticated actors can extract value from transaction ordering. Enshrining this separation is intended to make the process more transparent and equitable.

For dApp developers, validators, and node operators, the upgrade reshapes core assumptions about how Ethereum processes transactions and manages its state. The state growth controls in EIP-8037 in particular aim to keep node operation sustainable over the long term, preventing the network's data requirements from growing to a point that threatens decentralization by pricing out smaller operators.

What Comes Next

With devnets now running the full suite of planned EIPs, the immediate next milestone is the hardening of the codebase followed by deployment to public testnets. Once the upgrade proves stable on testnets through dual audit phases and cross-client testing, developers will be in a position to announce a mainnet fork date. The current expectation places that activation in the second half of 2026.

For now, the message from Ethereum's developers is one of steady, deliberate progress. Glamsterdam has reached its final development phase, the largest fork since the Merge is taking shape across the devnets, and the focus has shifted to hardening the code and preparing for public testnet deployment. As the core developers have emphasized throughout, the priority remains getting the upgrade right rather than rushing to hit a specific date, with the goal of setting Ethereum up for much more scaling in the years ahead.