Ethereum’s Glamsterdam Test Will Ask Validators to Choose How Fast Layer 1 Scales
• October 1, 2026 7:27 am • CommentsEthereum’s next scaling test will not flip a switch and instantly produce 200 million-gas blocks. It will ask validators to decide how aggressively they want to move in that direction.
The Ethereum Foundation has scheduled the Glamsterdam upgrade for the Sepolia testnet at 13:53:36 UTC on October 6, at epoch 353,024 and slot 11,296,768. Node operators need compatible execution and consensus clients before the fork, while Hoodi and mainnet activation dates remain undecided.
Stakers must also update their beacon-node and validator software.
That software update is only the first decision. Prysm 7.2.0 and Teku 26.9.1 support Glamsterdam, but both keep a 60 million gas preference unless validators explicitly configure a higher number.
Operators who want to propose blocks with a 200 million preference have to opt in. The Foundation’s release table lists six compatible execution clients, giving operators several supported routes into the test.
Glamsterdam is coming, and it is bringing blocks built without trusted relays (and a lot more)!
The Sepolia testnet upgrade has been scheduled for 6 October of this year. Hoodi and mainnet dates will be announced separately. Activation table, client releases, and info on the bug…
— stokes (@ralexstokes) September 28, 2026
A block gas limit sets how much aggregate computation can fit into a block. More room can support more transactions and demanding applications such as decentralized exchanges, especially during periods of heavy use.
It does not make Ethereum’s block time faster.
EIP-8261 introduces a schedule that helps clients coordinate recommended gas preferences at specified epochs while preserving validator choice. It does not make one gas limit a consensus-validity rule or invalidate a block merely because its proposer selected a different preference.
Blocks remain valid above or below the recommendation, and validators can choose their own preference.
That means Sepolia’s observed limit should change block by block as different validators propose blocks. The network will not jump from 60 million to 200 million at one predetermined second.
The rate of movement will reveal how willing operators are to test the larger capacity target. It will also show whether client defaults or operator choices have more influence during the public test.
The mechanism is intentionally flexible. A validator can keep a conservative preference while still following valid blocks proposed with a higher limit, allowing the network to test coordination without treating one setting as mandatory or forcing every operator to change configuration at once.
CryptoSlate notes that Ethereum has already climbed from 30 million gas toward 36 million in early 2025 and later to 45 million, while Fusaka client releases adopted 60 million as their default. Glamsterdam would make the next experiment considerably larger, testing more aggregate execution room without allowing any single transaction to consume an unlimited share of the block.
Glamsterdam combines the Amsterdam execution upgrade with Gloas on the consensus layer. Its headline changes include enshrined proposer-builder separation and block-level access lists.
The latter give clients better information about which accounts and storage locations a block touches, opening more room for parallel processing.
Those improvements are part of what makes a higher capacity target plausible. They do not eliminate the burden on node hardware or the risk of implementation bugs.
That is why Sepolia matters: it gives client teams and operators a public environment to test the new rules and the higher preference before any mainnet date is set.
Glamsterdam security vulnerabilities are now in scope for the Bug Bounty Program (https://t.co/MaBGlzTJ9S)
Come help us secure Ethereum to earn rewards and your place on the leaderboard alongside legends like @revofusion @mhswende and @GuidoVranken https://t.co/FhMz0YWLbL
— Fredrik (@0xfredrik) September 29, 2026
The Foundation has opened Glamsterdam specifications and EIPs to its bug-bounty program. Hoodi and mainnet activation dates remain undecided.
The signal to watch on October 6 is more than whether Sepolia forks successfully. Validator adoption of the 200 million preference and the testnet’s block-by-block response will provide the useful evidence.
Client stability as capacity expands will matter just as much. Glamsterdam’s scaling story will be written by operator behavior as much as protocol code.
Join the conversation!
We have no tolerance for comments containing violence, racism, profanity, vulgarity, doxing, or discourteous behavior. If a comment is spam, instead of replying to it please click the icon below and to the right of that comment. Thank you for partnering with us to maintain fruitful conversation.
