Ethereum Developers Back 200M Gas Limit for Glamsterdam

Ethereum’s 200M Gas Limit Depends on Repricing

Gas is the unit Ethereum uses to measure computation and data operations inside a block. A larger block gas limit can allow more aggregate work, but it can also strain the computers that validate the chain if costly operations are priced too cheaply.

That is the problem the three repricing proposals address. EIP-2780 breaks the flat 21,000-gas transaction base cost into explicit components designed to reflect the work a transaction performs. It preserves the headline cost for ordinary ETH transfers to existing accounts while treating other paths more precisely.

EIP-8037 adds separate metering for state creation, targeting an average state-growth rate of 120 GiB a year at a 150 million-gas reference limit. EIP-8038 would increase and separate charges for state access and account or storage writes as the chain’s state grows.

The final test results are significant because the presentation says they did not require a fresh specification release or client code changes. The planned 200 million setting is more than three times the 60 million limit cited by the Ethereum Foundation’s 2026 protocol update, which describes larger gas limits, block-level access lists and repricing as connected scaling work.

It does not mean a typical transfer suddenly becomes three times cheaper. The block limit measures total block capacity, while individual transaction costs depend on the gas schedule, demand and the activity being performed. Pricing state growth more directly can also make some operations costlier even as total throughput rises.

Glamsterdam Tests Move From Provisional to Final

The original ACDE agenda asked whether the gas-repricing numbers were final and listed the Glamsterdam devnet known as Platåberget. The post-call slides answered that question by describing the values in the latest release as final and within safety limits, while noting that specific benchmark scenarios would be forwarded to client teams.

That distinction is important for the upgrade’s status. A technical sign-off on benchmarked values narrows uncertainty around a design, but Ethereum upgrades are adopted through specifications, independent client software, public test networks and coordinated mainnet releases. No Glamsterdam mainnet date appears in the materials reviewed for this report.

The result also gives a sharper, nearer-term frame to the long-range Lean Ethereum roadmap. That earlier plan discussed an aspirational 10,000 transactions-per-second layer-1 goal. The 200 million gas-limit work is a measured capacity proposal with stated safety assumptions, not confirmation of that broader performance target.

Ethereum’s current roadmap ties Glamsterdam to scaling components including repricings, block-level access lists, further blob-parameter increases and ePBS, or enshrined proposer-builder separation. The combination is intended to increase capacity without simply shifting more unpriced workload or long-lived state onto node operators.

Hegota Proposals Face a September 10 Client Deadline

The same ACDE call addressed Hegota, the upgrade expected to follow Glamsterdam. The agenda says its proposed-for-inclusion list should be final in two weeks, proposals without a clear champion would be automatically declined, and clients were asked to publish preference lists by Sept. 10.

Several ideas were listed for consideration, including bytecode deduplication, lower transient-storage costs, a PAY opcode, linear memory costing and post-quantum ML-DSA verification precompiles. The agenda also says Frame Transactions had moved to considered-for-inclusion as a native-account-abstraction direction, while the exact implementation remained open.

That keeps the upgrade process separate from a completed promise to users. A proposal on the agenda is not an activated protocol feature, and a client preference list is not a finalized hard-fork specification. Ethereum’s earlier quantum-safe key-registry work illustrates the same staged approach: research and registration designs can precede a production migration by multiple forks.

The Crypto Fear & Greed Index read 29, or Fear, on Aug. 14. It is market context rather than evidence about protocol readiness, but it comes as developers are attaching a concrete capacity number to Glamsterdam rather than a price target.

The next checkpoints are client feedback on the forwarded benchmarks, Glamsterdam release and testnet milestones, and a public mainnet schedule. Until then, the verified change is the completion of the repricing sign-off and a 200 million-gas target the developers said it can safely support, not a live Ethereum gas-limit increase.