Ethereum Glamsterdam Upgrade Guide: Gas Limit Increase, ePBS, and What’s Next

 

Ethereum's next mainnet upgrade has reached public testnets. According to Crypto Briefing, the Sepolia testnet activated Glamsterdam on October 6, 2026 at 13:53:36 UTC, at epoch 353,024, opening an optional block gas limit of 200,000,000 against a previous ceiling near 60 million, more than a threefold increase in block processing budget. The qualifier matters: that figure is a testnet parameter. Mainnet gas limits have not changed, and no mainnet activation date has been set.

 

What draws attention here is not a single number but the two structural questions the upgrade touches, namely who builds a block and how quickly a block can be verified. Configuration already merged into the eth-clients repository schedules the second public testnet, Hoodi, to fork at epoch 132,352 on October 26, 2026 at 17:42:48 UTC, with the same 200 million gas target written into its schedule. Before that, core developers meet on October 15 to continue working through Glamsterdam and the follow-up fork Hegotá, which is the most direct window available for reading the mainnet timeline.

 

 

 

Testnets are live, mainnet is not scheduled. Sepolia forked on October 6 and the Hoodi activation config was merged on October 9, pointing to October 26. The stated target for mainnet is the fourth quarter of 2026, with no confirmed date.

 

200 million gas is a test target, not a mainnet commitment. Validators on Sepolia opt into the higher ceiling themselves, final mainnet parameters remain undecided, and any conclusion about mainnet capacity is premature.

 

The two headliners are ePBS and block-level access lists. EIP-7732 writes proposer-builder separation into the protocol, while EIP-7928 declares state access in advance so execution can be parallelized.

 

How the gas limit gets raised is changing too. EIP-8261 proposes an optional gas limit schedule in consensus layer config that sets a default target and a recommended maximum without altering any consensus rule.

 

The division of labour with Fusaka is clean. The Fusaka mainnet announcement shows that upgrade activated on December 3, 2025 and addressed data availability and blob scaling. Glamsterdam addresses execution efficiency and block production structure.

 

Three dates are worth marking. The consensus layer core devs call on October 15, the Hoodi activation on October 26, and the November 1 Last Call deadline attached to several core EIPs.

 

 

 

Glamsterdam merges two tracks, Amsterdam on the execution layer and Gloas on the consensus layer, and the portmanteau name reflects exactly that. Per Crypto Briefing's account, the Sepolia fork completed at the scheduled epoch and validators can signal whether to move toward the 200 million target. Core devs call chronicler Christine Kim notes in her recap of the execution layer call that the Sepolia upgrade went smoothly, that Hoodi is the second and final public testnet before mainnet, and that client releases were expected by October 16.

 

The gap between testnet and mainnet deserves to be stated plainly. Staking on a testnet costs effectively nothing, so validator composition, economic incentives and the cost of attacking all differ from mainnet. A testnet sustaining 200 million gas therefore does not establish that mainnet can safely adopt the same parameter. What actually needs validating is block propagation latency, the pace of state growth, and whether home stakers can keep up under load, and those readings accumulate over Hoodi and beyond rather than in a single fork event.

 

 

A detail easily lost in coverage is that Ethereum's gas limit is never hard-coded by a fork. Validators express a preference block by block, and under EIP-1559 the per-block adjustment is capped at one part in 1,024 of the previous block, so the ceiling moves gradually. EIP-8261 productizes that process by adding an optional gas limit schedule to consensus layer configuration, giving a default target and a recommended maximum for each effective epoch, with clients warning when an operator's setting exceeds the recommendation while still honouring the operator's choice. The proposal states explicitly that it changes no consensus rules, so blocks above or below the scheduled value remain valid. The example values in the spec, 60 million, 75 million and 90 million, illustrate the format rather than announce a mainnet plan.

 

 

 

The Fusaka announcement confirms activation on December 3, 2025 at 21:49:11 UTC at slot 13,164,544, built around PeerDAS in EIP-7594, which lets nodes verify blob availability by sampling rather than downloading everything. The same post sets out a staged path for blob parameters, starting at a target of 6 and a maximum of 9 per block, then rising through two blob parameter only forks to a target of 10 and maximum of 15, and later a target of 14 and maximum of 21. EIP-7935 in that same release set the recommended default gas limit at 60 million, which is where mainnet sits today.

 

That work widened the data pipe available to layer 2s and did comparatively little for layer 1 execution. The Ethereum Foundation's protocol priorities update for 2026 puts it directly: the gas limit has moved from an initial 30 million to 60 million, the first significant increase since 2021, and the 2026 goal is to push it toward and beyond 100 million, with block-level access lists named as the technical prerequisite. Before blocks can get bigger, clients have to be able to verify them faster.

 

 

That same official document targeted Glamsterdam for the first half of 2026. Public testing began in October instead, and the official target has slipped to the fourth quarter without a confirmed date. Ethereum's upgrade timelines have always been elastic, which reflects not weak execution but the reality that changes closer to the consensus core take longer to validate. Pricing that elasticity into expectations is more useful than tracking any rumoured date.

 

 

 

The EIP-7732 specification states the problem crisply: most proposers already outsource block construction to builders, and that exchange currently runs through trusted intermediaries. The proposal replaces that trust with protocol rules, so that honest proposers get paid and honest builders see their payloads reach the chain.

 

Mechanically, block validation splits into consensus and execution steps, separated logically and in time. A new in-protocol builder role commits to an execution payload and pays the proposer, while a subset of validators known as the Payload Timeliness Committee attests to whether the builder revealed the payload on time, without executing it in full. The stated benefits include shortening the work validators must complete before attesting, keeping consensus blocks smaller so they propagate faster, reducing reorg risk from blob transactions, and removing the dependence on trusted middleware. The proposal sits in Last Call with a deadline of November 1, 2026.

 

 

EIP-7928 is equally direct. Transactions resist parallel execution because clients cannot know in advance which accounts and storage slots each will touch. The proposal requires every block to carry a complete record of the accounts and storage locations accessed, along with their post-execution values, and a block whose list is incomplete or padded with spurious entries is invalid. With that record, clients can prefetch state from disk in parallel, validate transactions concurrently, compute the state root in parallel, and update state without re-executing. The spec's own analysis estimates that 60% to 80% of transactions touch disjoint storage slots, with post-transaction state diffs covering much of the remainder.

 

Read together, the design is coherent. Access lists keep larger blocks verifiable at speed, and ePBS ensures those larger blocks are produced inside a healthier market structure. Per the EIPs Insight tracker, 18 core EIPs are currently scheduled for inclusion alongside seven associated proposals, with the remainder covering transaction cost reform, higher calldata fees, and updated gas pricing for state access and creation.

 

 

 

Even if mainnet eventually adopts a higher ceiling, users will not experience a one-off collapse in fees. More layer 1 block space first absorbs the demand currently priced out of the market, and only then shows up as lower unit costs, while demand itself recovers as costs fall. The more realistic expectation is that layer 1 capacity and blob scaling work together, giving layer 2 settlement and complex layer 1 interactions more room at the same time. The live ETH price will not reprice around a testnet parameter, but the network's long-run carrying capacity does change the denominator in any valuation.

 

 

One misreading to avoid is that ePBS eliminates MEV. Extractable value arises from the economics of transaction ordering, and rearranging the scheduling mechanism does not remove that property. What changes is who carries the trust risk. Today the exchange between proposers and builders runs through relays and middleware that both sides must trust; once the proposal lands, protocol rules guarantee it. For validators that means fewer intermediaries and a clearer revenue path. For the building market it may reshape both the barrier to entry and the competitive balance.

 

Upgrades run on testnet time while prices run in real time. Track every move on the ETH spot market

 

 

 

Introducing a new role introduces a new attack surface. CoinCentral reported that consensus developer Potuz raised a concern on a September 17 developer call: because participating costs nothing on a testnet, an attacker could spin up thousands of builder identities, rotate them, bid aggressively and then withhold payloads, disrupting block auctions. No real funds would be at risk, but it could stall testnet traffic and prevent developers from evaluating the upgrade properly. The same report notes that some client protections only engage after several payloads go missing, and that developers want the ability to ban individual builders. Surfacing this on a public testnet is exactly what public testnets are for.

 

 

The mainnet date is the largest unknown. The target is the fourth quarter of 2026, but until Hoodi has been validated, any specific date is speculation. Mainnet gas parameters are equally open, since the 200 million figure belongs to Sepolia and Hoodi configuration. Several core EIPs also carry a Last Call deadline of November 1, so specifications can still change before then.

 

 

In the clean case, Hoodi activates on October 26 and runs without incident, core devs fix a mainnet epoch at a subsequent call, the upgrade lands within the year, and attention shifts to how quickly mainnet gas limits actually climb.

 

In the delayed case, Hoodi exposes issues in the builder market or in client implementations and mainnet slips into early 2027. Given the history of major Ethereum upgrades this is far from unlikely, and a delay of this kind signals a functioning process rather than a failing roadmap.

 

In the conservative case, the upgrade ships on time but mainnet gas limits rise far more slowly than testnet configuration suggests. This is arguably the most probable path, because validators vote block by block according to their own hardware and bandwidth, and the capacity of home stakers remains a constraint core developers have been unwilling to trade away.

 

 

The nearest marker is the consensus layer core devs call at 14:00 UTC on October 15. Per the agenda published in the ethereum/pm repository, the topics include Glamsterdam, Hegotá and improvements to the hard fork process itself. CoinDesk's week ahead also lists the call among the week's crypto events. Client releases follow around October 16, the Hoodi fork on October 26, and the EIP Last Call deadline on November 1. As for Hegotá, official language still places it later, with higher gas limits, native account abstraction and post-quantum readiness named as directions.

 

 

For James Mitchell, the figure most likely to be misread is 200 million. It is an optional testnet parameter that validators adopt block by block, and the mainnet number will be governed by a different constraint entirely, namely whether a node on ordinary hardware can process larger blocks without falling behind. The Ethereum Foundation's own stated goal of pushing toward and beyond 100 million is the more realistic anchor, and it tells the story better than the headline: moving from 60 million to 100 million is a climb, not a jump.

 

The second misreading concerns ePBS and MEV. Writing the builder role into the protocol changes the trust assumptions around middleware; it does not change where extractable value comes from. For traders, the question worth watching is not whether total MEV falls but how concentration in the building market evolves. Protocolization usually lowers barriers to entry while also sharpening the advantages of scale, and which force dominates will only be legible from the actual distribution of block production several months after mainnet.

 

The chain from technical variable to price variable also deserves discounting. The upgrade alters the network's long-run carrying capacity, which works on the denominator of a valuation rather than the numerator. ETH trades near $2,531 with a market capitalization around $309 billion, and that valuation reflects overall risk appetite, the rate environment and the revenue structure of the layer 2 ecosystem rather than any single testnet fork. Historically, the volatility clustered around major upgrades has come from expectation trading rather than fundamental repricing, with the real effects showing up over subsequent quarters through fee revenue and on-chain activity.

 

Three measures are worth tracking from here. Block propagation latency and missed block rates on Hoodi will define where the mainnet ceiling can sit. The actual slope of mainnet gas limit increases will be more honest than any official statement, because it is decided by validators voting block by block. And the number and share of active builders after ePBS goes live will reveal whether this structural change ultimately opens the market or concentrates it. Until those three datasets exist, treating Glamsterdam as engineering still under validation is sounder than treating it as a trading catalyst.

 

 

 

Glamsterdam is Ethereum's next mainnet upgrade after Fusaka, named by combining the execution layer fork Amsterdam with the consensus layer fork Gloas. Its two headline proposals are EIP-7732, which enshrines proposer-builder separation in the protocol, and EIP-7928, which introduces block-level access lists to make parallel execution possible. Per the EIPs Insight tracker, 18 core EIPs are currently scheduled for inclusion alongside seven associated proposals, with the overall goal of improving layer 1 execution efficiency and scalability.

 

 

No. The 200 million figure is an optional target on the Sepolia testnet, and the same value appears in the Hoodi configuration, but mainnet gas limits are unchanged and final mainnet parameters have not been decided. Ethereum's gas limit is set by validators voting block by block, with per-block adjustments capped under EIP-1559, so even after the upgrade ships any increase will be gradual. The Foundation's stated direction for 2026 is toward and beyond 100 million.

 

 

No date has been confirmed. The stated target is the fourth quarter of 2026, but core developers need Hoodi to validate first before fixing a mainnet activation epoch. Sepolia activated on October 6 and the Hoodi configuration points to October 26. The most direct window is the all core devs calls, with the consensus layer call on October 15 carrying Glamsterdam on its agenda, since mainnet epochs are typically agreed in that forum and then written into client releases.

 

 

No. Extractable value comes from the economics of transaction ordering, and changing the scheduling mechanism does not remove it. What EIP-7732 changes is the trust structure: today the exchange between proposers and builders depends on relays and other trusted middleware, while under the proposal protocol rules guarantee the outcome for honest participants on both sides. The specification also notes the design shortens the work validators must do before attesting, keeps consensus blocks smaller, and reduces reorg risk from blob transactions.

 

 

Because they are the prerequisite for raising the gas limit. EIP-7928 requires each block to carry a complete record of the accounts and storage locations accessed plus their post-execution values, and a block with an incomplete list is invalid. That record lets clients prefetch data, validate transactions and compute the state root in parallel. The spec estimates that 60% to 80% of transactions touch disjoint storage slots, which is what makes parallelization worthwhile. Without it, simply raising the block ceiling would make verification unsustainable.

 

 

They address different layers of the problem. Fusaka activated on December 3, 2025 around PeerDAS data availability sampling, with blob parameters stepping up from a maximum of 9 per block through subsequent parameter forks, which widened the data bandwidth available to layer 2s. Glamsterdam targets layer 1 execution efficiency and block production structure, creating the conditions for parallel execution and an in-protocol builder role. The two are complementary rather than alternatives.

 

 

There is no deterministic link between a technical upgrade and price. The upgrade changes long-run carrying capacity and fee structure, effects that typically take several quarters to show up in on-chain data, while short-term pricing responds more to risk appetite and the rate environment. Historically, volatility around major upgrades has reflected expectation trading rather than fundamental repricing, and profit taking after activation is common. Following the engineering and making trading decisions from it are two separate exercises.

 

 

The information above is provided for general market information and technical analysis only and does not constitute investment advice, financial advice, legal advice, tax advice or a recommendation to trade. Prices of crypto assets, equities and other related financial assets can fluctuate sharply, and past performance, technical indicators and on-chain data do not guarantee future results. The testnet parameters, proposal statuses, schedules and price figures cited here reflect publicly available information at the time of publication, core developer decisions can change at any time, and Ethereum's official channels and all core devs call outcomes should be treated as authoritative. Readers should conduct their own research and make decisions based on their own financial circumstances, investment objectives and risk tolerance, consulting a qualified professional where appropriate. The MEXC Crypto Pulse team accepts no liability for any direct or indirect loss arising from the use of this information.

 

 

James Mitchell specializes in technical analysis, market trends, and trading strategies for both Bitcoin and altcoins. Based in London, he has over 10 years of experience in financial markets. Before joining MEXC Learn, James worked as a senior analyst at a leading European investment firm, where he developed expertise in risk management and quantitative trading. His transition to cryptocurrency markets began in 2017, and he has since become recognized for his data-driven approach. He holds a Master's degree in Financial Economics from the London School of Economics. His analytical approach combines traditional technical analysis with on-chain metrics to provide readers with actionable insights.

 

Areas of Expertise: Technical Analysis, Market Trends and Cycles, Trading Strategies, Bitcoin and Altcoin Analysis, Risk Management.

 

 

 

Want the fastest access to MEXC's latest updates? Join our official Telegram group now!

MEXC Platform: Download App | Zero Fee Trading | Buy USDT Now

Join MEXC Community: X (Twitter) | Telegram | Discord

Get Insights: MEXC Learn | Terms of Service | MEXC Blog | MEXC News

Account Verification: Understand KYC | How to Complete KYC

External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)