Ethereum’s Pectra upgrade reshaped how validators manage balances. The next upgrade, Glamsterdam, will materially improve exit liquidity.
The proposals included in Glamsterdam address two areas of particular interest to institutional stakers: the architecture of block building, and the speed at which validators can exit large staked ETH positions. In this post, we break down the Ethereum Improvement Proposals (EIPs) that matter most to stakers, what each one changes in practice, and what you’ll need to do when Glamsterdam goes live.
What is Glamsterdam?
Glamsterdam consists of “Gloas” referring to changes on the Consensus Layer (CL), and “Amsterdam” to changes on the Execution Layer (EL).
The upgrade is outlined in EIP-7773 and encompasses a broad set of protocol improvements.
The Ethereum Improvement Proposals (EIPs):
EIP-7773: Glamsterdam Meta EIP:
CL:
- EIP-7732: Enshrined Proposer-Builder Separation [headliner] [SFI]
- EIP-7688: Forward compatible consensus data structures [CFI]
- EIP-8045: Exclude slashed validators from proposing [CFI]
- EIP-8061: Increase exit and consolidation churn [CFI]
EL:
- EIP-7928: Block-Level Access Lists [headliner] [SFI]
- EIP-7610: Revert creation in case of non-empty storage [CFI]
- EIP-7708: ETH transfers emit a log [SFI]
- EIP-7778: Block gas accounting without refunds [SFI]
- EIP-7843: SLOTNUM opcode [SFI]
- EIP-7954: Increase Maximum Contract size [SFI]
- EIP-7976: Increase Calldata Floor Cost [SFI]
- EIP-7981: Increase Access List Cost [SFI]
- EIP-7997: Deterministic Factory Predeploy [CFI]
- EIP-8024: Backward compatible SWAPN, DUPN, EXCHANGE [SFI]
- EIP-8037: State Creation Gas Cost Increase [SFI]
- EIP-8038: State-access gas cost increase [CFI]
- EIP-8246: Remove SELFDESTRUCT Burn [CFI]
- EIP-8282: Builder Execution Requests [CFI]
Networking:
- EIP-7975: eth/70 – partial block receipt lists
- EIP-8070: Sparse Blobpool
- EIP-8136: Cell-Level Deltas for Data Column Broadcast
- EIP-8159: eth/71 – Block Access List Exchange
- EIP-8189: snap/2 – BAL-Based State Healing
Informational:
Since last publishing this piece on May 5th, 2026, the following EIP has been removed from the Glamsterdam update:
CL:
When will Glamsterdam launch on mainnet?
Testing on Glamsterdam has begun but there is no clear timeline for the mainnet fork. A reasonable estimate is Q4 2026. Time estimates are best guesses at the time of writing. Actual mainnet launch date for Glamsterdam may vary greatly.
What should stakers know?
While many of these changes affect applications, network infrastructure, and gas economics, two EIPs most directly impact validators and institutional stakers.
Note that a prior version of this piece included EIP-8080: Let exits use the consolidation queue, but that EIP has since been ‘Declined for Inclusion’ (DFI), as noted above.
These two EIPs are: EIP-7732 and EIP-8061.
EIP-7732: Bringing proposer builder separation onchain
Today, validators who want to outsource block construction to specialized builders (and capture more of the MEV in each block) do so through MEV-boost, a sidecar piece of software that is connected to relays.
Validators trust relays to release payloads on time; builders trust them not to take advantage of their access to the payload.
What EIP-7732 changes: EIP-7732, known as enshrined Proposer-Builder Separation (ePBS), formalizes a new type of block builder directly into the Ethereum protocol. These onchain builders are permissionless — anyone meeting an ETH collateral requirement can participate — and they operate without the relay intermediary.
What stakers need to know: The existing MEV-boost and relay setup continues to function exactly as it does today — there is no forced migration. The new ePBS builder market is strictly opt-in. When implementation details are finalized, participation could appear as a configuration flag on the validator client. Stakers will opt into the new market as a whole — selecting individual ePBS builders is not expected to be an option, based on current specifications.
Why it matters for institutional stakers: A more competitive, trust-minimized builder market could be positive for the variable execution layer (EL) component of staking rewards. More competition among builders generally means better block optimization over time. The near-term impact will depend on adoption rates and how quickly the new market matures. Figment will monitor ePBS builder activity closely and will publish a dedicated deep dive on ePBS as implementation details are finalized ahead of the upgrade.
For more on how execution layer and consensus layer rewards differ and what drives variability in each, see Execution Layer Rewards vs. Consensus Layer Rewards: A Guide for Institutional Ethereum Stakers.
EIP-8061: Significantly faster exit processing
The exit churn limit determines how much ETH can exit the active set per epoch. Currently, that limit sits at 256 ETH, a conservative figure that, arguably, is more restrictive than necessary from a security perspective.
With the current exit churn limit, queues build up quickly in periods of high exit demand. Toward the end of 2025, Ethereum saw a significant spike in exit queue length as validators sought to exit a large amount of ETH related to a security incident.
What EIP-8061 changes: EIP-8061 ties the exit churn limit to the total amount of ETH currently staked, using a calculation similar to the pre-Pectra approach. Based on approximately 41.2M ETH staked, the revised formula produces an exit churn limit of roughly 1,258 ETH per epoch. That’s approximately 4.9x larger than the current limit.
Importantly, the churn limit will grow as total staked ETH grows, so the actual figure at the mainnet fork will likely be higher than today’s estimate.
A similar recalculation applies to the consolidation churn limit, increasing it from approximately 373 ETH per epoch to 629 ETH per epoch (a ~2.46x increase).
Why it matters for institutional stakers: For large institutional stakers, a significant exit can already take days — and that delay carries both direct and operational costs. More importantly, periods of high exit demand weigh on all stakers that need liquidity. This proposal will drastically decrease the exit queue for those wanting to unstake ETH.
Inceased Gas per block
Glamsterdam will also bring bigger blocks – moving from a gas limit of 60M per block to a gas limit floor of 200M per block. This change aligns with Ethereum’s priority to increase L1 capacity and came out of the interop meeting in Svalbard.
What does more gas mean for execution layer (EL) rewards and the burn? Unfortunately, the answer is not clear. Making more blockspace available could reduce the price of gas in the short to medium term. However, it is unclear what the total effect will be – will demand for blockspace absorb most/all of the increase and how much will the price of gas change?
As a reminder the base fee is burnt by Ethereum while priority fees are paid to proposers. So the answers to the questions above likely depend on which type of fee (base or priority) is being analyzed.
There was an increase in the gas limit from 45M to 60M around the Fusaka upgrade in December of last year, with the increase in gas limit happening around November 26th. Looking at average gas used, we can see that roughly all of the extra blockspace was used (average daily gas used increased by roughly ⅓, in line with the gas limit increase):
The change in the average gas price is much more difficult to determine, especially amongst the volatility:
PBS blocks paint a similarly tricky picture:
The magnitude of the change with Glamsterdam will be much larger, but it does not make the outcome any more clear.
The best that can be said is that the win with increased gas is more blockspace for users and builders, which is a win for Ethereum and the larger ecosystem.
What to Expect Before and After the Upgrade
Glamsterdam continues Ethereum’s methodical development trajectory, addressing trust assumptions in the block builder market, and the practical limitations of the current exit queue under high demand.
For institutional operators, significantly faster exit processing reduces one of the more operationally consequential risks in staking, the cost associated with longer time-to-liquidity.
To discuss how Glamsterdam may affect your ETH staking strategy, meet with our team. As a final note, until the fork is implemented on mainnet all of the above information should be treated as speculative.
