Ethereum core developers emerged from a week-long interoperability sprint above the Arctic Circle with an agreement set to reshape the network’s economics: a post-Glamsterdam gas limit floor of 200 million, a 3.3x jump from the current ceiling of 60 million. If the upgrade ships as planned in June 2026, Ethereum’s layer-1 could process more computation per block than at any point in its history.
The sprint – called Soldøgn Interop, held in Longyearbyen, Svalbard – was recapped by Ethereum Foundation protocol support lead Tim Beiko on May 2, 2026. Three distinct engineering workstreams converged during the week, each contributing a different piece of the capacity expansion.
Three Upgrades Working Together
The first workstream is enshrined Proposer-Builder Separation, or ePBS. Currently, the separation between those who propose blocks and those who build them is handled off-chain through MEV-Boost software. Glamsterdam moves that relationship on-chain, setting explicit deadlines for block construction, payload reveal, and attestations. That structural change gives the execution layer more predictable headroom. Timing-related stress under high load drops sharply.
The second is Block-Level Access Lists, specified in EIP-7928. This change allows clients to pre-fetch a block’s full read/write set before execution begins, enabling parallel execution of transactions and batched input/output operations. It also enables parallel state-root computation, meaning the network can do significantly more work per block without hitting sequential bottlenecks. That’s the key mechanism that allows the gas limit to increase from 60 million to 200 million – the network can handle it because execution happens in parallel rather than in strict sequence.
The third is EIP-8037, a repricing of state-writing operations. Writing new data to Ethereum’s global state carries a cost, but the existing pricing structure has become misaligned with actual resource consumption. EIP-8037 raises the gas cost for creating new state, which prevents the higher overall gas limit from translating into runaway state growth. The team chose a simpler fixed-cost model during the sprint, dropping an earlier active per-byte approach to make testing and benchmarking cleaner.
What a 200M Gas Limit Actually Means
The 200 million gas target is a floor, not a ceiling – validators and the community could push it higher over time once the changes are live. To put the number in context: a simple ETH transfer currently costs 21,000 gas. A complex DeFi interaction might use 200,000 to 500,000 gas. At a 200 million gas limit, a single Ethereum block could contain roughly 400 simple transfers in the time it could previously contain 120, or execute significantly more DeFi operations without congestion.
For end users, the direct consequence is fee compression. Ethereum base fees are a function of block demand relative to block capacity. Triple the capacity without tripling the demand and fees drop hard. In low-traffic periods, base fees could approach zero. Lido advisor Hasu, writing before the sprint results were published, described the 200 million gas target as “transformational” for Ethereum’s long-term competitiveness against both layer-2 networks and rival layer-1 chains.
Fee compression has a real consequence for validators: less burned ETH under EIP-1559. Supply dynamics shift. ETH’s deflationary narrative weakens unless demand grows in step with the new capacity.
Implications for Layer-2 Networks
Ethereum’s layer-2 system – Arbitrum, Optimism, Base, zkSync, and others – has grown substantially in part because layer-1 fees pushed activity off-chain. If Glamsterdam makes layer-1 cheap enough for routine transactions, some of that activity could migrate back.
That’s not necessarily bad for layer-2 projects; they can still offer additional features, faster finality for specific use cases, and application-specific execution environments. But the competitive dynamics shift. Base, which is built by Coinbase and carries significant stablecoin and DeFi volume, runs on Ethereum’s security layer regardless of whether transactions happen on L1 or L2. The cost savings flow back to users either way.
Smaller EIPs and What Didn’t Make the Cut
Several smaller changes were resolved at the sprint. EIP-8061, which increases churn limits for validator exits and consolidations, was confirmed as included in Glamsterdam. EIP-8080, which proposed routing exits through the consolidation queue, was declined – the additional complexity wasn’t considered worth the marginal improvement. EIP-8045, covering proposer duties within the look-ahead window, was scoped down to a narrower implementation.
The trimming of EIP-8080 and the reduction of EIP-8045 reflects a deliberate effort to keep Glamsterdam’s scope manageable. Ethereum upgrades have historically been delayed when the scope expands mid-cycle. With June 2026 as the target, developers appear committed to shipping what’s ready rather than including everything that’s technically interesting.
FAQ
what’s the Ethereum Glamsterdam upgrade? Glamsterdam is Ethereum’s next planned hard fork, targeting a June 2026 activation. Its centerpiece is a tripling of the network’s gas limit from 60 million to 200 million, enabled by three coordinated changes: enshrined Proposer-Builder Separation (ePBS), Block-Level Access Lists (EIP-7928), and a state-write repricing (EIP-8037).
Will Glamsterdam make Ethereum fees cheaper? Almost certainly in the short to medium term. A higher gas limit means more block space, which reduces competition for that space and pushes base fees down. How far fees fall depends on how much demand grows in parallel, but the directional effect is significant fee compression.
How does Glamsterdam affect Ethereum’s layer-2 networks? Layer-2 networks like Arbitrum, Base, and Optimism benefit from cheaper Ethereum as their settlement layer. Some routine activity may migrate back to L1 if fees are low enough, but L2 networks offer other advantages beyond cost that are unlikely to disappear.
Sources: Ethereum Foundation blog (Tim Beiko, May 2 2026), The Defiant, Cointribune, MarketScreener, BingX
