Ethereum (ETH) Glamsterdam Upgrade Goes Live on Sepolia to Verify 200 Million Gas Target

Glamsterdam Activates on Sepolia

Ethereum's next network upgrade, Glamsterdam, began running on the Sepolia testnet on Wednesday, October 7, opening the phase in which client teams verify the upgrade's target of 200 million gas per block. The step moves the upgrade out of specification work and onto a public testnet, the environment where client bugs, consensus mismatches and performance regressions surface for the first time. Sepolia runs the same client software as mainnet at smaller scale, which makes it the standard first stop for hard fork testing before wider coordination. An activation on a testnet changes nothing on mainnet by design; the point is to find failures where they cost a re-run rather than user funds. Gas is the unit behind every transaction's gas fee, and the block gas limit caps how much computation each block may contain; a 200 million gas target therefore implies a multiple of current capacity, a figure developers want proven under live load rather than asserted on paper. Rollups, token swaps and plain transfers all draw on the same allowance, so the ceiling is the base constraint every application inherits. Node operators install the updated software, validators attest under the new rules, and engineers read reorg behavior, attestation performance and memory consumption at the higher limit. If the higher-limit blocks fill cleanly, the target graduates to later testnets and eventually a mainnet schedule; if they do not, the failure modes appear on Sepolia first, where a reset takes hours instead of an emergency coordination call. The market has not treated the milestone as a catalyst: our live monitoring shows the Ethereum price down 5.5% over the past 24 hours, a move our Ethereum technical analysis coverage tracks against a broader pullback in digital assets. We follow each stage of the rollout in our Ethereum (ETH) tag hub.

Sepolia's activation is the first public run for an upgrade that follows Fusaka in Ethereum's release cadence. Glamsterdam's headline candidates under active discussion are enshrined proposer-builder separation, which would move the split between block proposers and block builders into the protocol itself, and a reduction in slot time that would shorten the wait between blocks. Both ideas aim at the same bottleneck, how blocks get built and how quickly the chain can finalize them, but neither ships on the strength of a specification alone. Neither change is presented as confirmed mainnet scope; what this milestone establishes is the performance envelope the upgrade must prove, above all the 200 million gas figure the current run is designed to test. The upgrade path now being exercised stretches back years; our guide to the Ethereum 2.0 upgrade covers the fork process that reshaped the network, and Glamsterdam follows the same staged logic of test, verify, then coordinate a mainnet switch. Throughput at that scale also changes economics beyond the base layer: Layer 2 networks settle their data back to Ethereum (ETH) and price their own fees off mainnet costs, so a proven 200 million gas limit expands the capacity rollups can draw on. Cheaper and larger settlement capacity also matters for applications that pay gas on every interaction, from automated market makers to gaming systems. What has not been disclosed is a mainnet timeline: no activation block or date exists yet, and coordination typically requires clean runs across the public testnets plus client release candidates before core developers commit to a schedule. The Sepolia run gives that coordination its first data point. Spot positioning has offered no applause: Ethereum ETFs logged a $201.9 million single-day outflow on Monday, with BlackRock's ETHA the largest contributor in the flow record published earlier this week.

The Safeguard Is Sequencing

For us, the significance sits in sequencing rather than speed. Putting Glamsterdam on Sepolia before any mainnet decision is the mitigation the protocol has applied to every fork since its move to proof of stake: a bug found on a testnet costs developers a re-run, while a bug found after mainnet activation costs users. The number to watch is the one developers are watching, sustained gas throughput at the 200 million target; if client teams report instability, the mainnet window slips and the headline candidates get re-sequenced. No activation block exists yet, and the client release notes are where that date will appear first.