Ethereum’s next attempt to expand base-layer throughput includes contracts and transactions that create persistent state, which would cost far more gas for new accounts, storage slots, and deployed bytecode.
The Ethereum Foundation said the candidate Glamsterdam schedule is designed to support roughly three times more base throughput by aligning gas charges with the network resources each operation consumes.
The two proposals are scheduled for Glamsterdam, yet EIP-8037 and EIP-8038 both retain formal Review status. Ethereum’s official roadmap plans the upgrade for the fourth quarter of 2026, with fixed Sepolia, Hoodi, and mainnet fork dates still unannounced.
That leaves builders a testing window before the candidate parameters and deployment schedule harden.
Higher throughput moves the bill to state creation
Every new account, storage slot, and byte of contract code expands the persistent state that nodes retain and serve. EIP-8037 says the state portion of a Geth database was about 390 GiB in January 2026.
After Ethereum’s gas limit rose from 30 million to 60 million, average new state created each day increased from roughly 105 MiB to 326 MiB, an annual pace of about 116 GiB.
The proposal takes that post-increase rate and scales it proportionally to a deliberately severe 200 million gas-limit scenario. The result is roughly 387 GiB of annual growth, enough to cross a cited 650 GiB performance threshold within a year from the 390 GiB starting point.
The EIP describes the earlier 30 million-to-60 million response as non-linear, making 387 GiB a motivating extrapolation rather than a measured forecast.
Its proposed control is a common cost of 1,530 gas for every byte of new state and a separate state-gas dimension. At a reference block limit of 150 million, that parameter targets average growth of 120 GiB a year.
The proposal lists 160 GiB a year as the worst case at a 200 million limit after repricing.
| State-creation operation | Current charge | Candidate state-gas charge |
|---|---|---|
| Create a new account | 25,000 | 183,600 |
| Create a new storage slot | 20,000 | 97,920 |
| Deploy 24 KiB of code plus a new account | 4,947,200 | 37,784,880 |
At the transaction level, users pay for both execution and state gas. The split preserves room for computational work while placing a separate ceiling on permanent additions to network state.
EIP-8038 tackles access to and writes of existing state, raising selected account and storage costs from client benchmarks conducted against a state snapshot comparable to mainnet in March 2026.
Together, the proposals use the roughly threefold throughput figure as an engineering support target. Activation would create headroom for higher limits rather than guarantee an immediate tripling of mainnet capacity.
A 929.7 million-transaction replay separates higher-limit fixes from code changes
The public repricing impact dashboard covers 929,731,274 transactions in 4 million blocks from Dec. 3, 2024, through June 15, 2026.
Each transaction was replayed independently against its canonical pre-transaction state under the existing schedule and one candidate schedule. Researchers first used the transaction’s original gas limit, then allowed a ceiling up to 10 times that limit.
The comparison covered success, gas consumption, logs, output, and execution traces.
Under EIP-8037, 174,473,898 transaction replays failed at their original limit but succeeded with more gas, while 2,687,652 entered the potentially broken group. Under EIP-8038, 84,708,228 were fixable with a higher limit, and 3,036,537 were potentially broken.
Repeated activity from a busy application can dominate the count, so the figures do not describe millions of separate contracts at risk.
The dashboard defines potentially broken as a baseline-successful transaction that the candidate schedule failed to rescue at the tested 10-times ceiling.
That group includes out-of-gas cases and transactions that reverted for another reason after the new costs changed execution behavior. It is a counterfactual risk classification: users, wallets, contracts, builders, and fee markets can adapt before mainnet.
The much larger fixable cohort chiefly shifts work to frontends, bundlers and infrastructure providers, which must submit limits that reflect the new schedule. The harder cohort exposes assumptions that more top-level gas cannot cure, including fixed 2,300-gas stipends, hardcoded gas forwarded to internal calls, logic that branches on gasleft() and presigned transactions with fixed limits.
A public outreach report prioritizes versions of the eth-infinitism ERC-4337 EntryPoint and related smart-account stacks, including ZeroDev and Alchemy. It also identifies recurring counterfactual failures involving Across, Socket/Bungee, CoW Protocol, and 0x.
Immutable contracts make the toughest cases operationally expensive. A durable response can require a new EntryPoint, account, factory, or validator implementation followed by user migration. Other systems may be repaired through routing, batching, or gas-accounting changes.
ERC-4337 bundlers also need to distinguish the two classes: better simulation and higher submitted limits can resolve fixable failures, while a validation guard or internal gas assumption still reverts when the outer transaction receives more gas.
Who must update before Glamsterdam
Wallets, RPC providers, indexers, node tooling, and gas estimators must incorporate the new rules, while frontends and bundlers need to stop relying on cached constants or limits calibrated to the old schedule.
Both EIPs require eth_estimateGas and related logic to account for the new costs. The Platåberget testnet announcement also warns that tools built around a hardcoded maximum gas limit or a single gas dimension will break.
A plain ETH transfer to an existing account can still use 21,000 gas. Sending value to an account that does not yet exist adds a state-creation charge at runtime, so software that assumes every transfer fits the old constant needs revision.
Regular users can keep familiar workflows if wallets and infrastructure update correctly. Meanwhile, developers carry the immediate burden of resimulating transactions, reviewing fixed stipends and internal call limits, and testing affected paths against the candidate schedule.
The Glamsterdam fork was scheduled for Aug. 20 and was reported live with the repricing schedule by Aug. 24. The published sequence moves from stable devnets to the long-lived Sepolia and Hoodi public testnets, then to mainnet, with the 2026 roadmap target still subject to testing.
Ethereum can price its way toward more throughput, but builders that encoded yesterday’s gas assumptions now have a limited window to prove their contracts and tools can survive tomorrow’s schedule.
