Ethereum Developers Lock In 200M Gas Limit and ePBS Progress at Arctic Interop Sprint
Longyearbyen, Svalbard | May 2, 2026
Just over 100 Ethereum core developers gathered last week in Longyearbyen, Svalbard, at 78°N latitude above the Arctic Circle, for the Soldøgn Interop: a week-long sprint to harden the Glamsterdam network upgrade, Ethereum's next major hard fork targeting the first half of 2026. The group reached consensus on a more than tripling of the network's gas limit, advanced a critical block-building reform to near-complete multi-client status, and locked final numbers on a long-overdue repricing of state storage costs. Soldøgn deliberately returned to the working devnet model pioneered at earlier sprints such as Amphora and Edelweiss, departing from the open coworking format used at Berlinterop.
Gas Limit: The Biggest Capacity Jump in Ethereum's History
The most consequential agreement to come out of Svalbard was a floor of 200 million gas per block post-Glamsterdam, up from the current ceiling of roughly 60 million. This is the largest single capacity expansion Ethereum has ever approved. The increase is made practical by EIP-7928, known as Block-Level Access Lists (BALs), which allows the network to declare in advance which accounts and storage slots a block will touch before execution begins. That pre-declaration enables parallel transaction processing, preventing a 200-million-gas block from becoming an unmanageable sequential workload for node operators.
Ethereum's DeFi ecosystem currently holds about $46 billion in total value locked, representing roughly 68% of all DeFi TVL across blockchains. ETH itself trades near $2,300 with a market cap of approximately $275 billion. The network's ability to sustain that dominance depends on keeping throughput competitive, and a 200-million-gas limit points toward a target of around 10,000 transactions per second on the base layer, according to projections from third-party analysts including Phemex, BingX, and QuickNode.
ePBS: Removing Trusted Middlemen From Block Production
The consensus-layer headline of Glamsterdam is EIP-7732, known as enshrined Proposer-Builder Separation (ePBS). Today, between 80 and 90 percent of Ethereum blocks are assembled by external block builders and routed through off-chain relay services. These relays are trusted, off-chain intermediaries with no protocol-level enforcement mechanism. ePBS removes that trust assumption by embedding the handoff between block proposers and builders directly into the protocol.
The mechanism works in two stages: a proposer commits to a builder's block header, then the builder reveals the full payload, with validator committees enforcing payment. One side effect is an extension of the effective block propagation window from roughly two seconds to roughly nine seconds, giving builders more time to optimise. At Soldøgn, developers achieved a 4-client execution layer by 3-client consensus layer devnet for ePBS by Tuesday, one client short of the full 4×4 target configuration, and the broader Glamsterdam devnet-2 reached stable multi-client status by Friday. Pre-Soldøgn, the Ethereum Foundation's own April checkpoint noted that ePBS was "proving to be trickier than anticipated."
Repricing and Governance Updates
EIP-8037, which addresses state-creation costs, shifted from a dynamic to a fixed pricing model during the week, with final numbers confirmed on bal-devnet-6. Alongside EIP-8038, which covers state-access costs, this repricing is meant to better align on-chain fees with the actual storage burden imposed on node operators. Misaligned costs have contributed to Ethereum's state bloat problem for years, and fixing them becomes more urgent as the gas limit rises.
On the governance side, Soldøgn formalised new coordination leads for Ethereum's All Core Developers calls. Nixo and Ansgar will co-lead ACDE (execution layer), Pari will serve as interim lead for ACDC (consensus layer), and Barnabas, Mario, and Danceratopz will rotate moderation duties for ACDT (testing). Two proposals did not make the cut: EIP-8080, which would have routed exits through the consolidation queue, was declined, and EIP-8237, a top-up sync architecture, was deferred to a longer-term roadmap slot. Among other changes confirmed in scope at Soldøgn are EIP-8061, EIP-8045, and EIP-7688; the Ethereum Foundation's full recap lists the complete set.
Regional Stakes: South Asia and Africa
The Glamsterdam upgrades carry direct relevance for users outside established crypto markets. South Asia recorded an estimated $300 billion in crypto transaction volume in the first seven months of 2025, an 80 percent year-on-year increase, with India leading global crypto adoption for the second consecutive year. Lower effective gas costs and higher base-layer throughput reduce friction for the DeFi, remittance, and stablecoin use cases that dominate retail activity in India, Pakistan, Bangladesh, and Sri Lanka. Devcon 8, Ethereum's flagship developer conference, is scheduled for Mumbai from November 3 to 6, 2026, arriving just after Glamsterdam is expected to ship and bringing the event to South Asia for the first time.
In sub-Saharan Africa, four countries now rank in the global top 20 for crypto adoption, up from two in 2024. Nigeria, Ethiopia, and Kenya in particular rely heavily on stablecoin-based payments and informal savings on Ethereum. For users in those markets who avoid layer-2 networks due to complexity or trust concerns, L1 cost reductions matter directly. The Ethereum Applications Guild has announced structured regional builder programs for Africa in 2026, a forward-looking commitment to expanding the developer ecosystem in these high-growth markets.
What Comes Next
Soldøgn also produced early prototypes for FOCIL (EIP-7805), a censorship-resistance mechanism slated for Hegotá, Ethereum's H2 2026 upgrade. Native account abstraction, which would enable features like gasless onboarding and social recovery wallets, was discussed at Soldøgn but remains in early design; developers explored requirements without advancing a specific proposal. Hegotá also names Verkle Trees as a major pillar: a restructuring of Ethereum's state storage projected to reduce the storage burden by approximately 90 percent and enable stateless clients, a scope that explains why the feature was deferred beyond Glamsterdam.