Ethereum's Glamsterdam hard fork, targeted for H1 2026, represents the network's most structurally significant upgrade since The Merge. The fork bundles two headliner proposals — Enshrined Proposer-Builder Separation (EIP-7732) and Block-Level Access Lists (EIP-7928) — alongside a gas repricing p...
"With Block Access List, we are getting all the state that changes from transaction to transaction" — removing "the biggest bottleneck we have." — Gabriel Trintinalia, Consensys/Besu Client Developer
Ethereum's Glamsterdam hard fork, targeted for H1 2026, represents the network's most structurally significant upgrade since The Merge. The fork bundles two headliner proposals — Enshrined Proposer-Builder Separation (EIP-7732) and Block-Level Access Lists (EIP-7928) — alongside a gas repricing package (EIP-7904) and 25+ additional EIPs under consideration. If delivered as designed, the upgrade paths Ethereum toward 10,000 TPS on the base layer, a 78% reduction in gas fees, and a gas limit increase from the current 60 million to 200 million per block.
Development is underway. The Ethereum Foundation's DevOps team has completed testing of three proposed EIPs on Devnet-4 and is transitioning to Devnet-5. Block-Level Access Lists already have active devnets; the more complex ePBS component has yet to reach its first devnet, with epbs-devnet-1 targeted for discussion at the April 16 All Core Developers Consensus call. Public testnets and dual audit phases are planned for spring 2026. Developers have stressed that correctness takes priority over the June target date, meaning a Q3 slip remains possible.
The upgrade's economic implications are substantial. It restructures how blocks are built on Ethereum, moving the $3-7B annual MEV economy from off-chain relay infrastructure into the protocol itself. It also lays groundwork for drastically cheaper L2 settlement. According to Tomasz Stańczak, former co-executive director of the Ethereum Foundation, the gas limit increase will be phased: 100 million initially, rising to 200 million once ePBS is fully operational. Further increases to 300 million may follow before year-end.
Ethereum's block production currently depends on off-chain infrastructure. Approximately 80-90% of blocks are produced through third-party relays — primarily Flashbots — that coordinate between block builders and validators. This system works but introduces trust assumptions, single points of failure, and centralization pressure. The top three block builders control more than 80% of PBS blocks.
EIP-7732 moves this coordination into the Ethereum protocol. Under ePBS, block builders assemble blocks and cryptographically seal their contents. Validators (proposers) select the highest-paying sealed block without visibility into its contents. Transactions are revealed only after finalization. A new Payload Timeliness Committee (PTC) validates payload delivery, replacing the relay trust model with protocol-enforced commitments.
The proposal was selected as the consensus-layer headliner for Glamsterdam during All Core Developers Consensus Call #162. According to Justin Drake of the Ethereum Foundation, approximately 10% of validators are expected to adopt ZK validation capabilities post-ePBS, unlocking additional throughput gains.
EIP-7928 addresses Ethereum's sequential processing bottleneck. Currently, the EVM processes transactions one after another because each transaction could modify state that the next one reads. This serial execution constrains throughput regardless of how much gas is available.
Block-Level Access Lists (BALs) pre-declare the accounts and storage slots accessed within a block. With explicit state mappings, disk I/O, EVM execution, and post-state root calculations become fully parallelizable. According to Toni Wahrstätter's analysis, the average BAL size runs approximately 35 KiB per block at 36 million gas, with worst-case sizes remaining below calldata worst cases.
The new mechanism introduces a block_access_list_root field in block headers, extends the Engine API to carry BALs in payloads, and creates a new eth/71 wire protocol for peer-to-peer BAL distribution. Execution layer clients are required to store BALs for a minimum of 3,533 epochs.
Together, the two headliners address Ethereum's primary architectural constraints: centralized block production (ePBS) and serial execution (BALs).
EIP-7904 anchors a package of gas repricing proposals that align opcode costs with empirical benchmarks. The Gas Cost Estimator project conducted extensive testing across seven EVM client implementations to measure actual computational effort for various opcodes. The findings revealed persistent misalignment between the gas cost schedule and real-world computational load.
The repricing targets 18 underperforming contracts benchmarked below 60 Mgas/s. Additional EIPs under consideration include:
| EIP | Function | |-----|----------| | EIP-7904 | General gas repricing based on empirical benchmarks | | EIP-8011 | Multidimensional gas metering (computation, storage, bandwidth) | | EIP-8032 | Size-based storage gas pricing | | EIP-8037 | State creation gas cost increase | | EIP-8038 | State-access (cold read) gas cost increase | | EIP-7954 | Increase maximum contract size limit | | EIP-7942 | Availability voting in validator attestations |
The net effect of the repricing package is projected at a 78% reduction in gas fees across both simple transfers and complex smart contract interactions. This figure applies to L1 transactions; L2 users stand to benefit additionally through cheaper blob posting.
Glamsterdam is in active development with progress advancing unevenly across its two headliners:
Block-Level Access Lists (EIP-7928): Already tested on multiple devnets. BAL-devnet-3 was targeted for launch in late March / early April 2026. The simpler design — essentially a structured data attachment to blocks — has allowed faster implementation across clients.
Enshrined PBS (EIP-7732): More complex and further behind. The first ePBS devnet (epbs-devnet-1) has yet to launch. Discussion is scheduled for the April 16 ACDC call. ePBS rewires the consensus layer's block production pipeline, requiring coordination across all consensus clients.
Overall status: Three EIPs tested on Devnet-4; transition to Devnet-5 underway. Public testnets and dual audit phases are planned for spring 2026.
The official target remains H1 2026, with June cited as aspirational. However, the former EF co-executive director Stańczak publicly cautioned the community, stating: "no amount of talking about Ethereum's roadmap and vision matters if we cannot achieve coordination levels that consistently meet goals on schedule." Given ePBS's complexity and the devnet gap, a Q3 2026 slip appears plausible.
A leadership transition adds uncertainty. Stańczak stepped down as co-executive director in February 2026. The Ethereum Foundation's new leadership must maintain coordination momentum across multiple client teams during the final development push.
Glamsterdam's MEV reform operates on two levels.
What it fixes: ePBS eliminates dependency on trusted relays, removes the "free option" problem (which affects approximately 0.82% of blocks on average, rising to 6% during volatile periods), and prevents proposers from tampering with block contents. By mid-2025, more than 50% of high-value Ethereum transactions were routed through private channels to avoid MEV extraction. ePBS addresses this by making block building protocol-native rather than relay-dependent.
What it does not fix: A January 2026 academic paper modeling ePBS in the presence of MEV found that while it reduces validator-side concentration, it "significantly amplifies profit and content centralisation" among builders. Access to private order flow still confers a structural bidding edge that compounds over time. Vitalik Buterin acknowledged this limitation, stating that "ePBS only prevents builder centralization from spilling over into the staking layer, but the issue of builder centralization itself remains."
The $3-7B annual MEV economy does not disappear under ePBS. It moves from off-chain relay infrastructure into the protocol, improving transparency and removing trust assumptions but preserving the underlying extraction dynamics.
Glamsterdam compounds the L2 fee reduction trajectory that began with the Dencun upgrade's EIP-4844 (blob transactions) in March 2024 and continued with Fusaka's expanded blob capacity.
With Glamsterdam's gas limit increase and potential expansion to 72+ blobs per block, rollups gain substantially more capacity to post compressed transaction data to Ethereum. Projections suggest a 60-95% reduction in average L2 transaction fees, potentially pushing per-transaction costs to fractions of a cent while maintaining Ethereum's security guarantees.
However, this creates a revenue paradox for Ethereum L1. In 2024, L2s paid $113 million to Ethereum for security and data availability. By 2025, this figure dropped 53% as blob usage became more efficient post-Dencun. Glamsterdam will accelerate this trend. Ethereum's L1 fee revenue — currently approximately $65 million annually — faces further compression as the network optimizes for rollup throughput rather than L1 fee extraction.
The economic logic requires Ethereum to trade direct fee revenue for network effects: cheaper L2 settlement attracts more rollups, which attract more users, which generates demand for blob space, which eventually reprices upward. Whether this flywheel materializes at the scale needed to offset L1 revenue decline remains an open question.
Viewed through the lens of blockchain economic sustainability, Glamsterdam addresses a structural problem but does not resolve it.
Ethereum currently operates on approximately $65 million in annual L1 fee revenue against $4-5B in annual staking inflation subsidies — a ratio that underscores continued dependency on token issuance rather than organic demand. Glamsterdam's throughput improvements could expand the fee base by enabling more transactions at lower unit cost, but the 78% fee reduction works against revenue growth in the near term.
The upgrade's real economic significance lies in infrastructure cost reduction. By eliminating relay dependency (ePBS) and enabling parallel execution (BALs), Glamsterdam reduces the operational overhead for validators and node operators. If the gas limit reaches 200 million, blocks carry 3.3x more computation — meaning the same security infrastructure processes substantially more economic activity.
From the Ethereum Foundation's January 2026 "Checkpoint #8" report, the organization's 2024 spending reached $134.5 million against a $1B treasury. The Foundation's shift to ETH staking and DeFi participation (announced in early 2025) partially addresses treasury sustainability but does not change the fundamental revenue equation: Ethereum's base layer generates approximately $65 million in fees annually while the ecosystem spends orders of magnitude more to maintain and develop it.
Glamsterdam is an engineering-driven upgrade with measurable targets: 3.3x gas limit expansion, parallel execution via BALs, protocol-native block building via ePBS, and empirically benchmarked gas repricing. The technical ambition is matched by execution risk — two major architectural changes to different protocol layers, developed concurrently, under a compressed timeline and a leadership transition.
The economic implications are mixed. Cheaper execution expands accessible throughput but compresses per-transaction revenue. The MEV reform improves transparency without eliminating extraction. The L2 cost reduction benefits end users but further erodes Ethereum's direct fee income.
Ethereum's bet with Glamsterdam is the same bet it has made since the rollup-centric roadmap: that cheaper infrastructure drives adoption at a rate that eventually generates self-sustaining fee revenue. At $65 million in annual L1 fees against a $4-5B staking subsidy, that threshold remains distant. But the fork moves the architecture materially closer to where it would need to be for that transition to occur.