Ethereum developers have tentatively agreed to activate the Glamsterdam upgrade on the Sepolia testnet on October 6 at 13:53 UTC, providing the clearest public-testing milestone yet for the network’s next major upgrade. The date replaces an earlier September 28 target discussed by core developers. However, October 6 remains tentative and depends on developers achieving a stable Glamsterdam devnet transition beforehand.
Ethereum’s official roadmap currently targets Glamsterdam for mainnet during the fourth quarter of 2026 but does not provide a specific date. That uncertainty is deliberate: developers intend to complete Sepolia and Hoodi testing before committing to mainnet activation.
Developers Want Another Stable Devnet First
The additional testing reflects the complexity of Glamsterdam. Its headline consensus change, enshrined proposer-builder separation, or ePBS, brings the relationship between validators and block builders directly into Ethereum’s protocol.
Today, validators commonly rely on external infrastructure when outsourcing block construction. ePBS moves that relationship into consensus itself, separating agreement over a block from its processing while reducing reliance on trusted third-party software.
That substantially changes Ethereum’s block-production pipeline. Earlier testing exposed issues including a builder-deposit caching bug that caused non-finality during a Glamsterdam devnet transition. Affected client teams subsequently developed fixes, but developers want another clean transition before advancing to Sepolia.
The Ethereum Foundation has already launched Platåberget, a longer-lived public testing environment specifically designed to expose applications and infrastructure to Glamsterdam before the upgrade reaches Sepolia and Hoodi. Developers are therefore treating October 6 as a target rather than a guaranteed fork date.
Glamsterdam Targets Higher L1 Capacity
Glamsterdam’s second headline feature is Block-Level Access Lists, or BALs. BALs identify which parts of Ethereum’s state a block accesses and modifies, allowing clients to understand dependencies before executing transactions. The architecture lays groundwork for parallel execution and disk reads, potentially allowing Ethereum to process substantially more activity without forcing node hardware requirements upward at the same rate.
Glamsterdam also includes a coordinated package of gas repricings aimed at supporting roughly a 200 million gas floor. Those changes affect applications directly. The Ethereum Foundation has warned that wallets, indexers and gas-estimation tools relying on hardcoded assumptions about Ethereum’s maximum gas limit will need to update.
Other changes increase the maximum deployed contract size from 24 KiB to 64 KiB and maximum initcode from 48 KiB to 128 KiB. The upgrade’s scope is now largely frozen, although Ethereum’s Glamsterdam meta-EIP remains in draft and individual components can still change before mainnet. The testing sequence explains why developers have not announced a mainnet date.
Ethereum’s process moves major upgrades through devnets before deploying them on long-lived public testnets. Sepolia is expected to come first, followed by Hoodi. Developers can then assess client compatibility, network finality and downstream infrastructure before choosing mainnet activation. A successful October 6 Sepolia fork would therefore represent an important milestone, but not the final hurdle.
The schedule also matters because Ethereum developers are simultaneously preparing Hegotá, the upgrade expected after Glamsterdam in 2027. Hegotá is currently expected to include fork-choice enforced inclusion lists and Frame Transactions, while Glamsterdam remains focused heavily on scaling Ethereum’s Layer 1 and restructuring block production.
For now, Q4 remains the official window for Glamsterdam. Whether Ethereum can meet it will depend on what happens before October 6—and, more importantly, whether Sepolia and Hoodi remain stable once Glamsterdam arrives.