Ethereum 2026 Upgrade Roadmap: Glamsterdam & Hegota Explained

Ethereum developers have confirmed a packed 2026 roadmap, anchored by two major network upgrades codenamed “Glamsterdam” and “Hegota.” Together, these hard forks are designed to make Ethereum faster, cheaper, harder to censor, and easier to run, while moving the project to a predictable twice‑yearly upgrade cadence. For users, builders, and investors, 2026 is shaping up as a turning point in Ethereum’s long‑term transition toward what many are calling “Ethereum 3.0.”

Beyond catchy names, these upgrades are tightly linked to the scaling and security strategy Ethereum has pursued since the Merge. With proto‑danksharding already live via the 2024 Cancun–Deneb (Dencun) upgrade and Pectra and Fusaka delivered in 2025, Glamsterdam and Hegota are the next steps in pushing throughput into the tens of thousands of transactions per second while keeping the network credibly neutral and decentralized.

From Annual Releases to a Biannual Upgrade Cycle

One of the biggest behind‑the‑scenes changes is not any single feature, but Ethereum’s new release rhythm. Core developers have committed to a biannual upgrade cycle, with Glamsterdam scheduled for the first half of 2026 and Hegota slated for late 2026. This is a deliberate shift away from slower, monolithic hard forks toward smaller, more frequent changes that are easier to ship and safer to review.

The move comes after criticism that Ethereum’s base‑layer innovation was lagging behind high‑throughput competitors, even as activity on its layer‑2 rollups exploded. A twice‑yearly cadence is meant to keep the protocol nimble: developers can respond faster to performance bottlenecks, security findings, and evolving use cases, while the community gains a predictable schedule to plan around.

Glamsterdam: Mid‑2026 Upgrade for Speed, MEV, and Gas Efficiency

Parallel Transaction Execution and Higher Gas Limits

Glamsterdam, expected in the first half of 2026, is the “speed and efficiency” fork. Its most eye‑catching promise is a leap in how many transactions Ethereum can process in parallel. Today, Ethereum executes transactions largely one by one, which leaves multi‑core hardware underused. Glamsterdam introduces Block Access Lists (EIP‑7928), a mechanism that pre‑maps which accounts and storage slots each transaction will touch.

With that information, clients can safely run non‑overlapping transactions simultaneously, exploiting multiple CPU cores without creating conflicts. Engineers at Consensys and other teams have described this as a way to remove long‑standing execution bottlenecks and unlock big performance gains without simply cranking up hardware requirements. In tandem, Ethereum’s gas limit is expected to increase in stages, potentially reaching around 200 million gas per block after Glamsterdam, up from roughly 60 million today.

For end users, the impact should be felt as shorter confirmation times and more headroom for on‑chain activity, particularly during peak demand. For DeFi and NFT protocols, parallel execution promises smoother performance under load, reducing the kind of congestion that has historically driven gas prices sharply higher in bull markets.

Enshrined Proposer‑Builder Separation and MEV Reform

A second pillar of Glamsterdam is Enshrined Proposer‑Builder Separation (ePBS), which effectively moves a widely used MEV‑Boost‑style design into the protocol itself. In simple terms, it formally separates block builders—entities that construct the most profitable block—from block proposers—the validators that ultimately sign and broadcast that block.

By integrating this separation at the protocol level, Ethereum aims to reduce centralization pressure around powerful block builders and mitigate censorship risks that arise when a small number of entities control which transactions make it on‑chain. ePBS also makes it safer for validators to participate in zero‑knowledge (ZK) proof‑based validation by giving them more time to verify proofs without risking missed slots or penalties.

In practice, ePBS should mean a more competitive and open market for block building, a fairer distribution of MEV revenue across validators, and a network that is more resistant to blacklisting or regulatory capture.

Boosting Layer‑2 Rollups and Data Capacity

Glamsterdam does not only focus on Layer 1. It also builds on proto‑danksharding to expand the number of data “blobs” available per block, potentially up to 72 or more. This extra blob space is critical for Ethereum’s rollup‑centric roadmap: it gives optimistic and ZK rollups more room to publish compressed transaction data, allowing them to handle hundreds of thousands of transactions per second at extremely low cost.

If this plays out as designed, users interacting with rollups could see fees drop to fractions of a cent, while maintaining Ethereum’s security guarantees. For Ethereum’s competitiveness against high‑throughput chains, cheaper and more scalable rollups may be the strongest answer to date.

Hegota (Heze‑Bogota): Late‑2026 Upgrade for State, Storage, and Privacy

What Is Hegota and Why It Matters

The second major 2026 upgrade, codenamed Hegota, is planned for late in the year. The name fuses “Bogota” (the execution‑layer fork) with “Heze” (the consensus‑layer fork), following Ethereum’s convention of linking Devcon host cities with astronomical names. Whereas Glamsterdam is about immediate performance and MEV structure, Hegota tackles longer‑term challenges around state bloat, node storage, and censorship resistance.

Developers are still refining the final feature set, with scoping discussions taking place on early‑2026 All Core Developers calls. However, several headline features are already emerging, with Verkle Trees and enhanced privacy and inclusion mechanisms at the top of the list.

Verkle Trees and the Path to Statelessness

One of Ethereum’s biggest long‑term bottlenecks is its rapidly growing state: the total data of accounts, contracts, and storage. Today, full nodes must store hundreds of gigabytes or more, making it increasingly difficult for everyday users to run their own infrastructure. Verkle trees, a more compact commitment scheme than the current Merkle Patricia trees, are Ethereum’s answer.

By dramatically shrinking proof sizes and enabling so‑called “stateless” or near‑stateless clients, Verkle trees could reduce storage requirements for node operators by up to an order of magnitude—down to just a few gigabytes in some projections. New nodes may be able to sync in minutes instead of days, significantly lowering the hardware and time barriers to joining the network.

For decentralization, this shift is crucial. If running a full node becomes feasible on consumer‑grade hardware again, more individuals and smaller organizations can independently verify the chain instead of relying on centralized RPC providers and exchanges.

Fork‑Choice Inclusion Lists and Censorship Resistance

Alongside state management, Hegota is expected to introduce features aimed at privacy and censorship resistance. One of the most discussed proposals is Fork‑Choice Inclusion Lists (FOCIL), which would allow groups of validators to collaboratively ensure that certain transactions are eventually included, even if parts of the network are unreliable or censoring.

Combined with Glamsterdam’s ePBS, these mechanisms are designed to protect Ethereum’s neutrality in an era of growing regulatory pressure. If implemented well, they would make it much harder for any single government or corporate actor to systematically block specific users or applications from the network.

Security, Institutions, and the Push Toward “Ethereum 3.0”

Underpinning the 2026 roadmap is a renewed emphasis on cryptographic strength. Researchers affiliated with the Ethereum Foundation have set a goal of achieving “128‑bit provable security” by the end of 2026, a benchmark seen as necessary for institutional‑grade financial use cases. This includes upgraded proof systems, audited implementations, and better tooling for ZK‑based rollups and clients.

For institutional players watching Ethereum’s evolution, the combination of higher throughput, lower fees, lighter nodes, and stronger formal security guarantees is a key narrative. Analysts argue that full danksharding, Verkle trees, and account‑abstraction improvements arriving across 2025–2027 could turn Ethereum into a backbone for global financial and Web3 infrastructure, capable of handling 100,000+ transactions per second across L2s with sub‑cent costs.

What It Means for Users, Builders, and Investors

For everyday users, the headline takeaway from Glamsterdam and Hegota is simple: faster and cheaper transactions on rollups, smoother performance on mainnet during busy periods, and a more censorship‑resistant base layer. DeFi, gaming, and NFT platforms built on L2s stand to benefit from higher blob capacity and lower data costs, while on‑chain actions that still rely on L1 should face less congestion.

For developers, the changes are deeper. Parallel execution and Block Access Lists will require careful thinking about transaction ordering, concurrency, and tooling. Verkle trees and stateless clients will change how applications reason about storage proofs and state access. Teams maintaining wallets, dapps, and infrastructure will need to adapt to new RPC patterns, witness formats, and validation flows.

For investors and market participants, Ethereum’s 2026 roadmap is a double‑edged signal. On one hand, it shows a mature protocol with a clear, research‑driven path to scale and security—potentially justifying long‑term bullish theses on ETH as a core settlement asset. On the other, it highlights ongoing execution risk: complex upgrades, coordination across clients, and the chance of delays or unforeseen bugs. How effectively Ethereum navigates this roadmap will shape not only its own valuation, but the trajectory of the broader Web3 ecosystem.

What is clear is that 2026 will not be just another incremental year for Ethereum. If Glamsterdam and Hegota land as planned, the network that emerges at the end of the year will be faster, lighter, and more resilient than the one that entered it—a critical milestone in the long journey to a truly global, decentralized computing platform.