Ethereum's next network upgrade, Glamsterdam, isn't ready for mainnet, and its Hoodi testnet rehearsal is on hold. The blocker is a gas change inside the release, EIP-8037, which reprices what it costs to create new state on the chain, and core developers want more evidence that existing mainnet contracts can survive it before they flip the switch.
As of the Ethereum Foundation's November 6, 2026 update, there is no Hoodi activation date and no Glamsterdam mainnet date. Everything else is clearer: what the upgrade bundles, why one piece of it paused the test run, and what, if anything, you need to do about it.
Key Facts
- Glamsterdam combines the Amsterdam execution-layer upgrade with the Gloas consensus-layer upgrade. Its headline changes are enshrined proposer-builder separation (EIP-7732) and block-level access lists (EIP-7928), per the Ethereum Foundation's testnet announcement.
- The upgrade activated on the Sepolia testnet on October 6, 2026, at 13:53:36 UTC, at epoch 353,024, slot 11,296,768, per the same announcement.
- Core developers delayed the Hoodi fork at ACDE #248 on October 22, 2026, four days before its scheduled October 26 activation, to study how EIP-8037 interacts with mainnet smart contracts, per the Hoodi delay post.
- On November 6, 2026, the Ethereum Foundation said AllCoreDevs would not fork Hoodi until those issues are resolved, in a later update.
- No action is required from Ethereum mainnet users or ETH holders because of the Hoodi delay.
The Small Step Glamsterdam Removes
Enshrined proposer-builder separation, or ePBS, is the change worth explaining slowly, because it removes a handoff that currently happens outside the protocol.
A block proposer and a block builder are separate jobs, but the way one hands a block to the other leans on third-party relays and off-protocol agreements. EIP-7732 moves that handoff into Ethereum's consensus protocol: a proposer includes a builder's commitment to an execution payload, and the builder subsequently reveals that payload.
The point of moving it is trust. ethereum.org's roadmap page says the change removes off-protocol trust assumptions and the reliance on third-party relays, and introduces trustless builder payments directly into the protocol. It also buys room: separating the two roles at the protocol level expands the propagation window, the time available to spread data across the network, from 2 seconds to about 9 seconds.
That extra time is what allows much larger payloads, which is how a block-building change becomes an L1 scaling change. A useful way to picture it is a relay that currently sits between builder and proposer, handing over the winning block; ePBS folds that handover into the rules of the chain itself. The comparison stops there, because ePBS changes consensus behaviour, not who can build.
What Block-Level Access Lists Change
Block-level access lists, or BALs, go after a different bottleneck: transactions getting validated one after another.
EIP-7928 introduces enforced lists that record the accounts and storage locations accessed during a block, together with post-transaction state changes. With that map in hand, clients can read state from disk in parallel, validate transactions in parallel, and compute state roots more efficiently. ethereum.org describes the effect as removing sequential processing bottlenecks by providing an upfront map of all transaction dependencies, setting the stage for parallel processing instead of one-by-one validation.
There is a networking piece to match. Block Access List Exchange (eth/71, or EIP-8159) is the direct companion to the lists and is now required for all execution layer clients, which is meant to enable faster syncing and executionless state updates. If you build on Ethereum, this is the part that touches your tooling most directly, even though it looks like plumbing.
The Gas Repricing That Paused Hoodi
Glamsterdam also rewrites parts of the gas schedule. EIP-8037 increases and separately meters the cost of creating state, and EIP-8038 updates state-access costs. The intent, per the Ethereum Foundation, is gas pricing that better reflects the cost of execution and state growth.
The scale matters here. Ethereum is scaling toward a 200M gas limit floor enabled by Glamsterdam, with developers testing at a 150M reference block gas limit to derive accurate state pricing, according to ethereum.org. At that size, the Foundation says state growth would become unsustainable without repricing it.
That repricing is what stopped the Hoodi rehearsal. Core developers decided at ACDE #248 on October 22, 2026, to delay the previously scheduled upgrade on Hoodi so they could investigate how EIP-8037 interacts with mainnet smart contracts. The same announcement removed the gas-limit schedule that would have raised Hoodi's gas limit to 200M at the same epoch.
The concern is specific. ChainSecurity identified cases where a call can fail and the contract that made it can still complete its failure handler, whereas under the earlier gas rules the same path might have caused the outer transaction to revert. Developers are checking assumptions across client implementations and existing contracts, with ChainSecurity, Dedaub and internal reviewers contributing.
The Foundation's November 6, 2026 update is worth quoting directly: "Based on these findings, ACD has decided to delay the hardfork process and not fork Hoodi until the issues are resolved." It also recommends that ACD wait "until the audit findings are public and there has been time to review them before deciding whether to revise the specifications or when to schedule the Hoodi upgrade." To help outsiders check their own code, the Foundation set up a repository listing eight invariants that will be broken by Glamsterdam.
What Developers Should Test, and What Users Can Ignore
Application developers should test contracts and gas estimation against the new Glamsterdam gas rules. Contracts that rely on fixed gas stipends, hardcoded gas limits, or assumptions about remaining gas may need changes. That is the practical to-do list; the timing is the open part.
For everyone else, the Hoodi postponement is quiet. The announcement concerns Hoodi, and no action is required from Ethereum mainnet users or ETH holders as a result of the update. The delay also does not set a Glamsterdam mainnet activation date.
The Dated Steps So Far
- September 17, 2026: Ethereum blogs the Glamsterdam testnet plan, naming the bundle and its headline changes.
- October 6, 2026, 13:53:36 UTC: Glamsterdam activates on Sepolia at epoch 353,024, slot 11,296,768.
- October 22, 2026: ACDE #248 delays the Hoodi activation and drops the 200M gas-limit schedule that would have come with it.
- October 26, 2026, 17:42:48 UTC: the original Hoodi slot, epoch 132,352, slot 4,235,264, passes without the fork.
- November 6, 2026: the Ethereum Foundation keeps Hoodi on hold until the review issues are resolved, and notes that work on Hegota continues in parallel.
Milestone
Network
Date
Status
Glamsterdam activation
Sepolia
October 6, 2026, 13:53:36 UTC
Live
Gas limit raised to 200M
Hoodi
Same epoch as the scheduled fork
Removed with the delay
Glamsterdam activation
Hoodi
October 26, 2026, 17:42:48 UTC (original)
Postponed
Glamsterdam activation
Hoodi
Not yet set
On hold
Glamsterdam activation
Ethereum mainnet
Q4 2026, date not confirmed
Not scheduled
Status as of November 6, 2026.
What Would Change the Answer, and When
Hoodi has no new date. ethereum.org lists mainnet deployment only as expected in Q4 2026 with the date not yet confirmed, and the Hoodi delay post explicitly does not set a mainnet activation date. The scope of Glamsterdam is frozen but can still change before mainnet, because the meta EIP remains in draft. The roadmap page reviewed also listed the Hoodi fork as the next milestone at October 27, 2026, a date the October 22 delay has since overtaken.
One caveat worth flagging: throughput and fee-reduction claims circulate in third-party explainers, but they do not appear in the Ethereum Foundation, ethereum.org or GitHub pages reviewed here, so treat them as unverified. The mechanics that are documented, ePBS, block-level access lists and the gas repricing, are the ones you can check against the primary sources above.
What would move this forward is specific and dated. Dedaub and ChainSecurity reports were due to be published on November 13, 2026, and their contents are not yet confirmed. A November 16 ACDT call and a November 19 ACDE call are the next scheduled checkpoints. A decision to revise the EIP-8037 and EIP-8038 specifications, or a new Hoodi activation epoch, would be the signal that the hold has lifted. Until then, the honest status is the one the Foundation gave: live on Sepolia, on hold for Hoodi, and not yet scheduled for mainnet.
