Ethereum’s Glamsterdam upgrade targets L1 scaling, with mainnet activation likely slipping to Q4 2026
Ethereum’s upcoming Glamsterdam upgrade is being described by core developers as the largest protocol-level overhaul since The Merge. The package combines an execution-layer track called Amsterdam with a consensus-layer track called Gloas, and follows the earlier Fusaka upgrade with a clear focus on layer-1 scaling. At its core, Glamsterdam changes how Ethereum processes transactions, builds blocks and manages a growing state database. The upgrade is built around three goals: faster processing through parallelization, higher throughput by splitting block building from validation work, and better long-term sustainability through fee changes that more closely reflect storage and hardware costs. Its two headline proposals are ePBS, or enshrined proposer-builder separation under EIP-7732, and BALs, or block access lists under EIP-7928. The first aims to move proposer-builder coordination into the protocol and extend the effective validation window from 2 seconds to about 9 seconds. The second would let the network identify non-conflicting transactions in advance and execute them in parallel, while also helping new nodes sync faster. Glamsterdam also includes repricing proposals for persistent state growth and data reads. As for timing, a previously cited schedule pointed to a Sept. 16, 2026 mainnet target, but more recent developments suggest a delay. According to the source article, Sepolia and Hoodi deployments are now expected to follow in September, pushing the mainnet goal into the fourth quarter of 2026, and possibly later in the year before a final ACD decision locks in a block height.






