I've been following Ethereum's upgrade roadmap for years, and I want to be direct about something: Glamsterdam is not a routine maintenance release. The developers working on it are describing it as the most significant protocol change since Ethereum moved to proof‑of‑stake in 2022. And today, it officially entered its final development phase.
Ethereum core developers confirmed they are now running developer networks, called devnets, containing the complete set of Ethereum Improvement Proposals planned for Glamsterdam. This is the last closed testing environment before the codebase gets hardened and deployed to public testnets. After that comes mainnet.
No firm launch date has been set yet, but the current target is the second half of 2026.
What "Largest Fork Since the Merge" Actually Means
Parithosh Jayanthi, a core developer and devops engineer at the Ethereum Foundation, put it plainly. He described Glamsterdam as probably the largest fork Ethereum has had since the Merge, adding that it will change a lot of assumptions about the network and set it up for significantly more scaling going forward.
That's not hype language from a marketing deck. That's a technical lead describing what the codebase actually does. So let me break down the two headline proposals, because they matter.
Enshrined Proposer‑Builder Separation Is the Big One
The first major change is enshrined Proposer‑Builder Separation, formally tracked as EIP‑7732. Right now, the process of building transaction blocks and proposing them to the network is split between different actors, but that separation happens offchain, through infrastructure that introduces trust assumptions and creates centralisation pressure.
Glamsterdam brings that separation directly into Ethereum's core protocol. By moving it onchain, developers aim to reduce the manipulation opportunities that arise from maximal extractable value, or MEV, the extra profit validators and searchers can extract by controlling the order of transactions in a block. This has been one of Ethereum's most persistent fairness problems. ePBS is the most serious structural attempt to address it at the protocol level.
Block‑Level Access Lists Speed Things Up
The second headline feature is Block‑level Access Lists, tracked as EIP‑7928. This proposal allows blocks to declare in advance which accounts and smart‑contract data they plan to access during execution. Ethereum clients can then preload that information ahead of time rather than fetching it reactively mid‑execution.
The practical result is faster and more predictable block processing, an important step for making Ethereum easier to scale using zero‑knowledge proving systems, which are increasingly central to Ethereum's long‑term scaling roadmap.
Gas Repricings Will Change How Much Things Cost
Beyond the two flagship proposals, Glamsterdam includes a broad set of gas repricings that will meaningfully change the economics of transacting on Ethereum. Jayanthi described the direction clearly: high‑level compute operations get cheaper, while state‑related operations become more expensive.
This is designed to make Ethereum's fee structure more accurately reflect actual resource consumption and to make the network more compatible with ZK proving technology. It's also the kind of change that will require significant outreach to the developer community, which is exactly what the team is doing now.
Where Things Stand
Developers are currently focused on three things: continued testing on the devnets, finalising the technical specifications for each EIP, and communicating the repricing implications to the broader Ethereum ecosystem so builders can plan ahead.
The testnets come next, and public testing will be the first time the broader community gets hands‑on experience with what Glamsterdam actually does to the network. The progress being reported from inside the team sounds genuine and steady.
Glamsterdam is coming. And when it arrives, Ethereum will look and behave quite differently from the network it is today.
