Glamsterdam and Hegota forks, L1 scale

The coming year is set to be critical for the Ethereum scale. In the year In 2026, the Glamsterdam fork will bring fully parallel processing to the chain and increase the gas limit to 200 million, up from 60 million today.

Substantial validators switch from re-enacting transactions to zero-knowledge (ZK) validators. This sets Ethereum Layer 1 on a path to 10,000 transactions per second (TPS) and beyond, although the target won't be hit until 2026.

Meanwhile, data blobs increase (perhaps as many as 72 or more per block), allowing Layer 2s (L2s) to process hundreds of thousands of transactions per second. L2s are also becoming easier to use; ZKsync's latest Atlas update allows funds to remain on the mainnet but can be quickly traded in an on-chain execution environment within ZKsync's Elastic Network.

The proposed Ethereum interoperability layer will enable seamless cross-chain operation between L2s, privacy will take center stage, and improved censorship resistance will be targeted for the Heze-Bogota fork later this year.

Ethereum in 2026: The Glumsterdam Fork

Ethereum developers are finalizing what should be included in the Ethereum Improvement Proposals (EIPs) in the Glumsterdam Hard Fork, which is expected in mid-2026. Confirmed header changes are block access lists and included proposer separators. Neither looks particularly exciting, but they have the potential to populate the blockchain before turning to ZK Tech.

At some point, Core Divis will come up with cool names for things like “Firedance,” but until then, we'll be stuck with the boring technical names they chose.

Glamsterdam: block access lists (EIP-7928)

Although “block access lists” sounds like a censorship scheme, the update actually enables “perfect” parallel blocking processing.

To date, Ethereum has been operating in single-line mode, with a very long queue of transactions executed sequentially, one after the other. Block access lists allow scaling to a multi-lane highway, where multiple transactions are processed simultaneously.

The term refers to the mapping included in each block, designed by the block producer, who first runs everything on some high-end hardware. The map tells Ethereum clients which transactions affect which other transactions, accounts and storage locations, and what the state differences are after the transaction. This allows them to harvest the transactions and run them simultaneously on multiple CPU cores without conflict.

Related: Blockchains Quietly Prepare for Quantum Risk as Bitcoin Timeline Debates

“With a block access list, we're getting all the conditions that change from transaction to transaction, and we're putting that information in the block,” explained Gabriel Tritinalia, a senior blockchain engineer at Consensus who is working on it for the forced client.

It also allows customers to preload all the necessary data from disk to memory, rather than having to keep reading the disk sequentially, which Trintinalia calls “the biggest bottleneck we have.”

Fully parallel processing allows Ethereum to process high transactions per second and have large block sizes without increasing its gas limit.

Glamsterdam: A proven proponent of separation

The process of identifying block builders and proponents is initiated by MEV Boost, a non-protocol solution that uses a centralized relay as an intermediary and identifies approximately 90% of the blocks. Enshrined Proposer Builder Separation (ePBS) integrates this process directly with Ethereum's consensus layer for reliable operation.

The idea of ​​separating the two is that block builders compete with each other to select and order transactions in the best possible way to build a block, while suppliers choose which block to choose. The aim is to reduce the maximum extractable value (MEV) centralization pressure and improve security, decentralization and censorship resistance.

But in terms of efficiency, the main advantage of EPBS is that it provides more time to generate and distribute ZK-proofs throughout the network. Validators are currently penalized for being slow, which discourages waiting to validate ZK-verifiers. EPBS provides additional time to receive and verify ZK certifications.

This could give validators more time to receive proofs (and potentially more time to generate proofs), Ethereum researcher Ladislaus von Daniels said, adding that EPBS decouples prevent proof from executing executions and thus sends another flavor of delayed executions.

“This opt-in makes zkAttesting even more incentivized for certifications.”

Justin Drake, a researcher at the Ethereum Foundation, said that approximately 10% of validators will switch to ZK after this point, which will allow for a further increase in the gas limit.

Ethereum L1 gas addition and L2 blob target improvements

The gas limit (related to the supply of resources on L1) has already increased to 60 million. It should increase significantly by 2026 – although there are varying estimates of how much it will go.

“I think in 2026, I would expect to see 100 million fairly soon. Anything beyond that is probably just too speculative to consider,” said Gary Schulte, senior staff blockchain protocol engineer on the Besu client. He added that switching to a slow kill could result in a higher gas limit.

Related: Ethereum Triples Gas Limit ‘Floor, We Can Go Up' – Sassano

Ethereum Foundation co-director Tomas Stanchak predicted at a recent unbanked summit that the cap will increase to 100 million in the first half of 2026, and to 200 million following ePBS. Further improvements mean that up to 300 million gas per block is possible before the end of the year.

Ethereum creator Vitalik Buterin was more of an observer. At the end of November, he said that he “expects continued growth but more targeted/consistent growth for the next year. For example, one possible future: 5x increase in gas limits and 5x increase in gas costs for relatively inefficient operations. Buterin mentioned contracts with large contract sizes such as storage, precompiles and calls.”

Ethereum 2026 fork number 2: Heze-Bogotá

Expect to see some EIPs taken from Glamsterdam in this fork, but according to Forecast, the only EIP currently on the list for inclusion is the Fork-Select Inclusion List (FOCIL). This was for Glamsterdam but was pushed after heated debate, as it would require too much work and make life too difficult.

It's not focused on scale, but on cypherpunk's anti-censorship by encouraging multiple validators to include certain transactions in each block.

“This is a censorship protection mechanism, which ensures that if you have at least part of the network, if you are honest … then you have to include your transaction in a certain period of time,” said Trintinalia.

Stay tuned for Part 2 as we bring L1 to scale with ZK-proofs in 2026.

Magazine: The Big Questions: Will Bitcoin Survive a 10-Year Blackout?