Ethereum developers are moving through the final devnet phase of Glamsterdam, the network’s next major protocol upgrade and, by most client teams’ own description, its largest structural change since the 2022 Merge. Two Ethereum Improvement Proposals are confirmed for inclusion — EIP-7732 and EIP-7928 — with roughly ten additional EIPs scheduled alongside them.
The two confirmed EIPs
EIP-7732 (enshrined proposer-builder separation, or ePBS) takes a mechanism that already exists in practice — separating the validator that proposes a block from the entity that builds it — and writes it directly into the protocol. Today, that separation runs through off-chain relay infrastructure such as MEV-Boost. Enshrining it removes the dependency on trusted third-party relays and standardizes how proposers and builders interact at the protocol level.
EIP-7928 (block-level access lists, or BAL) requires every block to declare in advance which accounts and storage slots its transactions will read or write. Clients can then process non-conflicting transactions in parallel and pre-fetch the state they need, rather than executing everything sequentially. This is one of the more direct routes to higher effective throughput without simply raising the gas limit and hoping clients keep up.
What else is bundled in
A June 2026 tracking EIP (EIP-7773) listed ten proposals as scheduled for inclusion alongside the two headline changes, including EIP-7708 (a new structure for ETH transfer logs), EIP-7954 (raising the smart contract size limit from 24 KiB to 64 KiB), and EIP-8037 (finalized in May 2026, which sets a fixed cost per byte of new state and separates out a dedicated gas budget for state growth). None of these individually reshapes the user experience the way ePBS or BAL does, but together they touch contract deployment limits, log formats, and how the protocol prices state bloat — a persistent long-run concern for node operators.
Where testing stands
As of early August 2026, the upgrade is in its final devnet phase, with client teams testing a 200 million gas-limit floor as part of that process. Reaching a stable devnet result at that gas limit is a prerequisite for scheduling a public testnet and, eventually, a mainnet activation date — it does not by itself confirm one.
Timeline: aspirational versus likely
Developer documentation has referenced the second half of 2026 as the target window, with some earlier internal references to a mid-2026 date now widely treated as aspirational within client teams. Coordinated mainnet upgrades on Ethereum routinely slip by a calendar quarter once devnet and testnet issues surface, and Glamsterdam’s own tracking has already shown that pattern. Treat any specific month named before a public testnet activates as a planning target, not a commitment.
Performance claims deserve the same scrutiny
Coverage of Glamsterdam has cited throughput projections around 10,000 transactions per second under the combined effect of BAL-enabled parallelization and the higher gas-limit floor. That figure describes a theoretical ceiling under favorable conditions — client implementation quality, real-world transaction mix, and network propagation all affect what throughput actually looks like once the upgrade is live. Historical Ethereum upgrades have delivered real but more modest throughput gains than early projections suggested; there’s no reason to assume Glamsterdam breaks that pattern by default.
What this means for different users
Validators and node operators are the most directly affected group. ePBS changes how block building integrates with the protocol, and BAL changes what a block-processing pipeline has to do to take advantage of parallel execution. Operators running client software should track official client-team release notes rather than third-party summaries once a public testnet date is set.
Application developers building on Ethereum should watch EIP-7954’s larger contract size limit and EIP-8037’s separate state-growth gas budget, both of which affect how contracts are designed and priced going forward.
Everyday holders and traders are unlikely to notice a discrete before/after moment. Gas-fee effects, if any, will show up gradually and unevenly across transaction types rather than as a uniform drop the day the upgrade activates.
Bottom line
Glamsterdam bundles a structural change to how Ethereum blocks are proposed and built (ePBS) with a mechanism aimed at real parallel transaction processing (BAL), backed by a handful of smaller EIPs addressing contract limits and state growth. The devnet phase testing a 200 million gas-limit target in early August 2026 is genuine progress, but no mainnet date is locked in. Anyone tracking this for trading or infrastructure decisions should follow the public testnet announcement — that is the point at which “aspirational” timelines start becoming reliable ones.
Advertisement
Sources and review
This article was checked against the primary or authoritative sources below .
- What Is Ethereum Glamsterdam? 2026 Upgrade, ePBS, and Timeline — Tapbit
- Ethereum's Glamsterdam Upgrade Enters Final Devnet Phase With 200M Gas-Limit Target — The Defiant
- Ethereum Glamsterdam Upgrade & EIPs Explained — Datawallet
- Glamsterdam — ethereum.org
- Ethereum Glamsterdam Upgrade 2026: What Changes and Why ETH Traders Should Care — Phemex
Frequently asked questions
Glamsterdam is Ethereum's next major protocol upgrade, targeted for the second half of 2026. It bundles two confirmed Ethereum Improvement Proposals — EIP-7732 (enshrined proposer-builder separation) and EIP-7928 (block-level access lists) — alongside a set of smaller EIPs scheduled for inclusion.
Proposer-builder separation already exists off-protocol today through relays such as MEV-Boost. EIP-7732 moves that separation into the core protocol itself, so validators can rely on a standardized, enshrined mechanism for splitting block proposal from block building rather than depending on external relay infrastructure.
EIP-7928 requires each block to declare upfront which accounts and storage slots its transactions will touch. That lets clients process independent transactions in parallel and pre-fetch state, instead of executing transactions strictly one after another.
There is no confirmed mainnet date. Developer references have pointed to the second half of 2026, with the upgrade in its final devnet phase as of early August 2026 testing a 200 million gas-limit target. Client teams have described earlier timelines as aspirational, and the schedule could still shift to Q3 or Q4.
Not uniformly. The upgrade targets higher throughput and lower costs for specific transaction types tied to parallel execution and state growth, not a flat fee cut across every transaction. Projected throughput gains toward roughly 10,000 TPS are targets tied to specific conditions, not guaranteed outcomes for every user.
Advertisement