Ethereum's most consequential upgrade since The Merge is now on the launchpad. Glamsterdam, targeted for May–June 2026, bundles eight Ethereum Improvement Proposals (EIPs) that collectively aim to cut gas fees by up to 78%, triple the network's gas throughput ceiling to 200 million, and hard-wire...
"In Glamsterdam, Ethereum is getting ePBS, which lets proposers outsource to a free permissionless market of block builders. This ensures that block builder centralization does not creep into staking centralization, but it leaves the question: what do we do about block builder centralization?" — Vitalik Buterin, Ethereum Co-Founder, March 2, 2026
Ethereum's most consequential upgrade since The Merge is now on the launchpad. Glamsterdam, targeted for May–June 2026, bundles eight Ethereum Improvement Proposals (EIPs) that collectively aim to cut gas fees by up to 78%, triple the network's gas throughput ceiling to 200 million, and hard-wire anti-censorship protections directly into the consensus layer. For an ecosystem nursing six consecutive red monthly candles and an ETH price still struggling to hold $2,000, the stakes could not be higher.
The upgrade arrives at a moment of acute contradiction. Ethereum's Layer 1 gas fees have already collapsed to 0.185 gwei — effectively sub-penny transactions — yet the network's economic engine is sputtering. Daily fee revenue has cratered from $23 million at peak to $6.3 million, ETH burn has fallen 71% post-Pectra to roughly 3.26 ETH per day, and the "ultrasound money" narrative has given way to a modest 0.23% annual inflation rate. Glamsterdam is Ethereum's bet that radically cheaper and faster execution at Layer 1 will reignite demand, recapture activity that migrated to rollups and competing chains, and restore the economic gravity that once made holding ETH a deflationary proposition.
But the upgrade also confronts a structural crisis that most retail participants never see: the block building pipeline has become a two-player oligopoly, with Titan and Beaverbuild historically controlling over 90% of block production through exclusive order flow deals. Glamsterdam's flagship EIP-7732 — enshrined Proposer-Builder Separation — is designed to break that chokehold. Whether it succeeds or merely shifts the centralization vector will determine whether Ethereum can credibly claim to be the neutral settlement layer for a multi-trillion-dollar on-chain economy.
Glamsterdam is not a single feature — it is a coordinated overhaul spanning both Ethereum's execution and consensus layers. At its core are two headliner EIPs that reshape how blocks are built, validated, and processed.
EIP-7732: Enshrined Proposer-Builder Separation (ePBS)
Today, Ethereum's block production relies on an off-chain relay infrastructure dominated by MEV-Boost. Block builders assemble transaction bundles optimized for maximal extractable value, then submit them through relays to proposers who select the highest-paying block. This system works but concentrates immense power in the hands of a few sophisticated builders and the relay operators they depend on.
EIP-7732 moves this entire mechanism on-chain. Under ePBS, builders cryptographically commit to block contents through a sealed-bid process. Proposers select the highest bid without seeing — or being able to manipulate — what's inside. Transactions are revealed only after finalization. This eliminates the need for trusted relay intermediaries and standardizes MEV handoff rules at the protocol level.
EIP-7928: Block-Level Access Lists (BALs)
The second headliner is less visible but equally transformative. BALs require each block to declare its read/write footprint upfront — which accounts and storage slots it will touch. This seemingly simple change unlocks parallel transaction processing. Instead of the current sequential execution model, Ethereum clients can validate independent transactions simultaneously, dramatically increasing throughput.
EIP-7904: Benchmarked Gas Repricing
This proposal realigns gas costs with actual computational resource consumption. The result is a projected 78.6% reduction in gas fees for both simple transfers and complex smart contract interactions. Combined with a planned gas limit increase from the current ~60 million to a target of 200 million, this effectively creates a 10x expansion in Layer 1 capacity.
Additional EIPs in the bundle address blob data expansion (further reducing rollup posting costs), ZK proof verification improvements, and preliminary quantum resistance measures. The full scope of all eight proposals is still being finalized as devnet testing progresses.
The urgency behind EIP-7732 becomes clear when you examine the current block building landscape. According to data from relayscan.io, Titan currently controls approximately 50% of Ethereum blocks, with BuilderNet (Flashbots' decentralized alternative) at roughly 28%. Historically, Titan and Beaverbuild jointly controlled over 90% of all MEV-Boost blocks.
This concentration didn't emerge from superior technology alone. It was built on exclusive order flow agreements. Titan secured an exclusive deal with the Banana Gun trading bot in early 2024. Beaverbuild partnered with CoW Swap. These arrangements create a self-reinforcing cycle: more exclusive order flow means more profitable blocks, which means higher bids to proposers, which means more blocks won, which attracts more exclusive order flow.
The economic stakes are enormous. MEV extraction on Ethereum runs at approximately $40–60 million per month, with cumulative extraction surpassing $1.8 billion by mid-2025 and total annual MEV activity estimated at over $3 billion. Titan alone has generated roughly $19.7 million in total profit from block building.
Vitalik Buterin's March 2, 2026 blog post laid out a multi-layered defense:
The combination of ePBS, FOCIL, and encrypted mempools represents Buterin's vision of transforming Ethereum's base layer into a "neutral zone" where fairness is enforced by protocol mechanics rather than the benevolence of centralized infrastructure operators.
Glamsterdam's economic implications create a paradox that ETH holders must grapple with.
Current State (Pre-Glamsterdam): | Metric | Value | |---|---| | Average Gas Price | 0.185 gwei | | Daily Fee Revenue | ~$6.3M | | Daily ETH Burned | ~3.26 ETH | | Annual Inflation Rate | ~0.23% | | ETH Staked | 35.86M ETH (28.9% of supply) | | Active Validators | ~1.1 million | | ETH Price (Mar 4, 2026) | ~$2,082 | | Market Cap | ~$246B |
Ethereum's fee revenue already sits at cyclical lows. Gas at 0.185 gwei means a standard transfer costs fractions of a cent. The EIP-1559 burn mechanism — once burning 10,200 ETH daily — now barely registers. Ethereum has drifted from deflationary "ultrasound money" to a mildly inflationary asset.
The Glamsterdam Bet: By cutting gas costs another 78% and tripling throughput capacity, Glamsterdam wagers that radically cheaper execution will generate enough new demand volume to offset per-transaction fee compression. The playbook mirrors Amazon's strategy of cutting margins to grow total addressable market. If a 3x gas limit expansion attracts 5x the transaction volume, total fee revenue — and ETH burn — could increase despite lower unit costs.
However, the risk is real. If demand fails to materialize at the Layer 1 level — if users and applications remain satisfied with Layer 2 execution — then Glamsterdam could further erode Ethereum's fee revenue and accelerate ETH's drift toward structural inflation. The validator entry queue (3.4 million ETH waiting, ~60-day activation times) suggests institutional confidence remains strong, but that conviction will be tested.
Glamsterdam creates an existential question for Ethereum's rollup ecosystem: if L1 gas becomes cheap enough, do you still need a rollup?
Post-Dencun (2024), Layer 2 fees fell 70–95%. The January 2026 BPO update raised blob limits from 6 to 14 per block, further cheapening rollup data posting. Glamsterdam expands blobs again and raises the overall gas ceiling, which should reduce L2 costs by an additional 50–70% through 2026.
This is positive for rollups in the short term — cheaper data availability means better margins. But at 200 million gas and parallel execution, Ethereum L1 begins to compete directly with its own rollups on throughput and cost. Base, Arbitrum, and Optimism have built their value propositions around being "cheap Ethereum." If Ethereum itself becomes cheap, their competitive moat narrows to execution speed and application-specific features.
The likely outcome is a new equilibrium where rollups specialize rather than generalize. Application-specific rollups (gaming, social, high-frequency DeFi) will retain clear advantages. General-purpose rollups will need to differentiate on features — privacy, custom execution environments, chain abstraction — rather than raw cost.
As of early March 2026, the Ethereum Foundation's DevOps team has tested three EIPs on Devnet-4 and is transitioning to Devnet-5. Parallel testing is underway on bals-devnet-2 (for Block-Level Access Lists) and epbs-devnet-0 (for enshrined Proposer-Builder Separation). These devnets are critical for ensuring that major client implementations — Geth, Besu, Prysm, Lighthouse, and others — can interoperate seamlessly.
The official target remains May–June 2026, though the Ethereum Foundation has described this as "aspirational pending validation of critical components." Given Ethereum's history of upgrade delays (Pectra itself shipped several months later than originally planned), a Q3 slip would not surprise veteran observers.
The risk factors are nontrivial. ePBS is one of the most complex consensus changes since The Merge. Parallel execution via BALs requires significant client-side optimization. And the gas limit increase to 200 million cannot be activated until the new execution model proves stable under real-world conditions — likely a phased rollout rather than a single jump.
Glamsterdam is a three-part revolution: ePBS restructures block production, BALs enable parallel execution, and gas repricing slashes costs by up to 78%. Together, they target a 200 million gas ceiling — a 3.3x increase from today.
The block builder oligopoly is the core target. Two entities historically controlled 90%+ of Ethereum block production. EIP-7732 plus FOCIL and encrypted mempools aim to break this concentration at the protocol level.
The economic paradox is real. Cheaper gas could reignite demand and restore ETH burn, or it could further compress fee revenue. The outcome depends entirely on whether volume growth outpaces unit price decline.
Layer 2 rollups face an identity crisis. A radically cheaper L1 undermines the core value proposition of general-purpose rollups, forcing a pivot toward specialization.
Timeline risk is material. A May–June target is aspirational. ePBS and BALs are among the most complex changes Ethereum has attempted. Devnet testing is progressing but not yet on public testnets.
Validator conviction remains strong. With 35.86M ETH staked, 1.1 million active validators, and a 60-day entry queue, institutional commitment to Ethereum's long-term architecture is evident despite six consecutive red months on ETH's price chart.
Glamsterdam is not just a technical upgrade — it is Ethereum's answer to the fundamental question of whether a decentralized base layer can compete on performance without sacrificing neutrality. The block builder oligopoly represents exactly the kind of economic centralization that Ethereum was designed to prevent, and the combination of ePBS, FOCIL, and encrypted mempools is the most aggressive protocol-level intervention the network has ever attempted against entrenched intermediaries.
The economic stakes are clear. At $2,082 per ETH and a $246 billion market cap, Ethereum is priced for recovery but has yet to deliver the demand catalyst that would justify it. Glamsterdam's 200 million gas target and 78% fee reduction are designed to be that catalyst — a fundamental repricing of Layer 1 execution that could pull activity back from rollups, competing chains, and off-chain venues.
But hope is not a strategy. The upgrade's success will ultimately be measured not in technical benchmarks but in economic outcomes: Does fee revenue stabilize or grow? Does ETH burn recover? Does the block building market actually decentralize, or do sophisticated actors find new concentration vectors within the ePBS framework? The answers will arrive in the months following activation — and they will determine whether Ethereum's claim to be the world's neutral settlement layer is vindicated or revealed as aspiration.