Just over one hundred core contributors to the Ethereum network recently convened above the Arctic Circle in Longyearbyen, Svalbard, for the Soldégn Interop event. This intensive, week-long gathering focused on the critical Glamsterdam network upgrade, aiming to enhance Ethereum’s scalability and robustness. The Soldégn Interop followed last year’s Berlinterop but returned to the highly effective, single-track format previously utilized for significant upgrades like Amphora, Edelweiss, and Nyota. The primary objective of this focused approach was to drive multi-client progress toward the specific goals of hardening Glamsterdam.
By the conclusion of the week, the assembled developers achieved three paramount objectives: securing alignment on a post-Glamsterdam gas limit floor of 200 million, establishing stable implementations of Proposer-Builder Separation (PBS) operating with external builders, and finalizing the repricing numbers for EIP-8037. Beyond these core deliverables, substantial headway was also made on future features for the Hegotía upgrade, including advancements in FOCIL (Flashbots Optimization Chain Integration Layer) and native account abstraction, alongside discussions on a multitude of other technical topics.
The Unconventional Venue: Why Svalbard?
The choice of Longyearbyen, Svalbard, as the meeting location was deliberate and imbued with symbolic significance. Svalbard is a unique geopolitical anomaly, offering a rare environment where individuals of any nationality can reside and work without requiring a visa. This accessibility fosters an environment of global collaboration. More profoundly, the region is home to the Global Seed Vault and the Arctic World Archive, two state-of-the-art cold-storage facilities tunneled deep into the permafrost. These archives safeguard humanity’s most precious data, from agricultural backups to cultural artifacts and crucial source code, preserving them for potential future needs a millennium from now. Notably, a snapshot of Ethereum’s source code is already housed within the Arctic World Archive, underscoring the network’s commitment to long-term data integrity and preservation.
Furthermore, Svalbard experiences the phenomenon of the midnight sun from late April through August, meaning the sun never sets. This perpetual daylight mirrors the always-on nature of the Ethereum network, a characteristic that core developers aimed to embody during their intense week of work. The 24/7 uptime of the sun provided a fitting backdrop for a community dedicated to maintaining and enhancing a decentralized network that operates continuously.
Hardening Glamsterdam: Paving the Way for Scalability
The central ambition of the Soldégn Interop was to solidify the Glamsterdam upgrade and establish a target for the post-upgrade gas limit floor. Safely increasing the gas limit is a complex, multi-faceted challenge that Glamsterdam is designed to address. It tackles issues related to how blocks are constructed and proposed, the headroom available for client implementations under load, and how state creation costs scale in tandem with increased transaction throughput.
In practical terms, the week’s intensive efforts culminated in a stable, multi-client Glamsterdam development network. This network was running the latest versions of ePBS, incorporating updated block access list specifications, and generating crucial benchmarking data. This data serves as the foundation for proposing a credible and safe increase to the gas limit.
The majority of the developers’ time was dedicated to hands-on coding, often extending into the early morning hours. These intense coding sessions were punctuated by focused breakout groups designed to achieve consensus on design decisions and to deliberate on longer-term roadmap items. This immersive environment allowed for rapid iteration and problem-solving, a hallmark of successful interop events.
Supporting this effort, three Ethereum Foundation (EF) teams provided essential infrastructure. EthPandaOps deployed ethIQ, a performance monitoring tool, and a Panda Message Queue (MQ) server to facilitate agentic workflows. The Protocol Support team established soldogn.xyz as the definitive source of truth for interop goals, schedules, and documentation. The EF Digital Studio team documented the entire week’s proceedings, promising the release of the first-ever interop documentary.
The Pillars of Glamsterdam: ePBS, BALs, and Gas Repricings
The Glamsterdam upgrade rests on several key technical advancements, with ePBS, Block-Level Access Lists (BALs), and gas repricings being the most significant.
Enhanced Proposer-Builder Separation (ePBS)
ePBS represents a fundamental shift in how Ethereum blocks are constructed and proposed. Beyond refining the relationship between proposers and builders, ePBS restructures the slot by introducing explicit deadlines for block construction, payload reveal, and attestations. This introduces a greater degree of determinism in block production, allocating more time for the execution layer. This increased time allocation is critical for safely raising the gas limit.
The week’s ePBS efforts began with an ambitious goal: establishing a stable 4 Execution Layer (EL) and 4 Consensus Layer (CL) Glamsterdam development network by Monday evening. Initial challenges pushed this target to Tuesday, when a 4×3 configuration proved stable enough to commence stress testing. The remainder of the week was dedicated to an ePBS hardening cycle: stress testing, identifying and exposing edge cases, fixing issues, and repeating the process.
A significant breakthrough occurred during a Tuesday morning Builder API breakout session, which substantially simplified the specification for validator registration, the bid/header/commitments flow, the builder payment trust model, and circuit-breaker behavior. Mid-week debugging efforts zeroed in on cross-client edge cases, particularly concerning the invalidation of beacon requests by execution requests. A newly developed test suite revealed a gap in implementation across all participating clients. By Thursday morning, CL teams reported stable ePBS operation, while EL-side bid pathways were still undergoing debugging, with resolutions achieved by Friday. Two contentious points remain for AllCoreDevs (ACD): whether a request signature should commit to the receiving builder, and how to ensure a 1 ETH-staked-builder design remains resilient against Private Information Retrieval (PIR) Sybil-based liveness attacks. By Friday, nearly all clients were running in concert on glamsterdam-devnet-2, with the external builder pipeline tested end-to-end.
Block-Level Access Lists (BALs) and Gas Repricings
If ePBS is the consensus-layer component of Glamsterdam’s scaling strategy, the execution layer advancements are equally crucial. These primarily consist of gas repricings and the implementation of Block-Level Access Lists (BALs). By providing clients with upfront information about a block’s read and write sets, BALs enable parallel execution, batched I/O operations, and parallel state-root computation. These capabilities are fundamental to determining how large a block clients can efficiently handle.
The BAL track at Soldégn operated on separate development networks, distinct from the Glamsterdam ePBS chains. This isolation ensured that optimization benchmarks were not entangled with the complexities of consensus-layer stabilization. Each optimization was implemented behind its own feature flag, allowing for isolated measurement and comparison rather than evaluating them as a single, combined package. The BAL benchmark dashboard and leaderboard provided a clear view of each client’s worst-case performance across the test suite. By prioritizing the improvement of the slowest execution paths, the developers aimed to raise the gas limit floor across the board, not just for the most performant implementations.
A core element of Glamsterdam’s gas repricings is EIP-8037, which increases the gas cost for state creation. This measure is designed to prevent unbounded state growth, ensuring that a higher gas limit does not lead to a proportional increase in the amount of data stored on the blockchain. Prior to Soldégn, the EIP-8037 specification included dynamic per-state-byte pricing that was directly tied to the block gas limit. This made testing combinatorially challenging, requiring a separate fuzz matrix for each gas limit band, and rendered benchmarking nearly intractable. During the early stages of the Soldégn Interop, the teams agreed to abandon dynamic pricing in favor of a fixed cost_per_state_byte. Future repricing adjustments will be handled at fork boundaries rather than within a given fork.
The accounting model itself underwent an iterative development process. A Monday breakout session moved state-gas accounting from the end of the execution to the end of the call frame. A subsequent Tuesday session addressed account creation costs, code deposit costs, and CREATE-transaction reverts. On Wednesday, edge cases related to reservoir refunding and refilling necessitated a reevaluation of the model. A Thursday breakout session reverted the accounting to the opcode level, concluding that the primary complexity resided within the reservoir model, not the accounting computation itself. By Friday, the specification had stabilized on bal-devnet-6, and the BAL track had successfully delivered the final repricing numbers. This entire process highlights the critical role of interop events in resolving complex specification, implementation, testing, debugging, and design issues within hours and days, rather than weeks or months.
The convergence of these three threads—ePBS, BAL optimizations, and gas repricings—led to the week’s headline achievement: a credible 200 million gas limit floor for the post-Glamsterdam era. This significant increase is achievable because ePBS optimizes slot structure for execution time, BAL optimizations provide clients with the necessary throughput headroom, and EIP-8037 ensures that the higher gas limit does not result in uncontrolled state expansion.
Progress on Other Glamsterdam Threads
Beyond the core advancements of ePBS, BALs, and gas repricings, numerous other Glamsterdam-related EIPs and technical aspects were addressed during breakout sessions.
In the Consensus Layer (CL) domain, several smaller Glamsterdam EIPs saw their final decisions made. EIP-8061, which proposes an increase in exit and consolidation churn, was successfully integrated into glamsterdam-devnet-1. Conversely, EIP-8080, related to exits via the consolidation queue, was declined for inclusion in this upgrade. EIP-8045, concerning the removal of slashed validator duties, was scoped down to apply only to proposer duties within the look-ahead window. EIP-7688, focusing on SSZ stable containers, remains within the Glamsterdam scope but was intentionally held out of glamsterdam-devnet-1 while developers worked through the challenges of bounded gossip message sizes for attestations under progressive lists.
A critical architecture breakout session on Wednesday morning, involving both EL and CL teams, resulted in the deferral of EIP-8237 out of Glamsterdam. This decision was made to preserve optionality for a more extensive "top-up sync" architecture in a future fork. In its place, the participants agreed to draft a new EIP that would standardize the sequencing of forkchoiceUpdated, newPayload, and getPayload calls. It will also specify a snap-sync initiation handshake and tighten the consistency of valid/invalid states between the engine API surfaces.
Hardening and testing remained a pervasive theme throughout the week. A dedicated Thursday session addressed fork-choice compliance testing frameworks. The diamond repository, a collection of reproducible CL edge-case scenarios, was highlighted. Additionally, PandaOps’s external builder testing tool, buildoor, was demonstrated mid-session, with attendees actively suggesting attack scenarios in real-time.
Looking Beyond Glamsterdam: Hegotía and Future Developments
The Soldégn Interop also served as a crucial forum for discussing future upgrades, particularly Hegotía and subsequent forks. Several breakout sessions were dedicated to exploring these next steps.
A deliberately proposal-agnostic session focused on native Account Abstraction (AA). This discussion delved into the fundamental requirements and constraints that any future AA design must satisfy. The feature-set goals included support for alternative signature schemes, aggregation, batching, recovery mechanisms, gas sponsorship, flexible nonces, and keystore wallets. These were considered alongside hard constraints such as public mempool compatibility, statelessness, and Layer 2 Denial-of-Service (DoS) resistance.
A dedicated FOCIL breakout on Thursday concentrated on implementation updates. Early prototypes for FOCIL are already functional, with multi-client interop and a dedicated FOCIL development network identified as immediate next steps. Two significant design decisions were finalized: FOCIL will be disabled during periods of two-epoch non-finality, mirroring the behavior of the proposer-boost circuit breaker, and an index-based bookmark approach will be adopted to ensure compatibility with frame transactions and EIP-7702.
Further into the future, a long-running ETH Peer-to-Peer (P2P) track explored the development of a QUIC-based replacement for libp2p, designed with privacy-by-default features and slot-aware integration. Prototypes for erasure-coded broadcasts demonstrated a simulated propagation speed approximately six times faster than GossipSub on 2.4 MB payloads. The CL track also revealed a strong sentiment towards eventually deprecating consolidations entirely. The proposed approach involves declaring a final fork that supports them, followed by a mandatory exit-then-redeposit mechanism, which is seen as a cleaner long-term solution to validator-set state growth.
Refining the AllCoreDevs Process
On Wednesday afternoon, Nixo and Ansgar, the two AllCoreDevs (ACD) co-leads, convened a session to gather input from core contributors regarding the ACD process itself. The discussion revisited the "headliner" construct, a framework for selecting major upgrade themes. The pros and cons of maintaining a "strawmap," a community-driven roadmap, were debated, and the criteria for EIP Streamlined Fork Inclusion (SFI) were formalized. The consensus was to retain headliners but to relax the rigidity between EIPs and themes, accepting a "theme plus candidate EIP" as a viable pattern. The straw map’s per-fork year assignments beyond 2026 were identified as overly canonicalized and likely to be softened. A new four-point SFI definition was proposed, with the AllCoreDevs Technical (ACDT) group signaling readiness and the AllCoreDevs Execution (ACDE) and AllCoreDevs Consensus (ACDC) teams retaining the final decision-making authority. A new prioritization and ordering process, to be established after Consensus-critical Fork Inclusion (CFI) decisions and reflected in a meta-EIP, will replace SFI’s previous role in driving devnet inclusion, commencing with the Hegotía upgrade.
In terms of call coordination, Alex Stokes announced a three-month sabbatical commencing the following week. Pari will assume ACDC moderation responsibilities during this interim period, and Barnabas will cover for the ACDT. The current leadership structure for AllCoreDevs is as follows: Nixo and Ansgar chair ACDE, Pari is the interim ACDC moderator, and Mario, Barnabas, and Danceratopz rotate ACDT moderation duties.
Broadening the Scope: Miscellaneous Progress
In addition to the major technical advancements, developers used the in-person time to make progress on a wide array of other topics. This included improvements to test harnesses, notably compressing Hive feedback loops from hours to mere minutes. Enhancements were also made to engine API plumbing, including gossip deduplication, batched calls, and light-client-driven head discovery. Significant hard tradeoffs were discussed concerning client diversity, alongside numerous other critical development areas. A comprehensive list of all session notes is publicly available at soldogn.xyz.
Next Steps and Future Outlook
Following the productive Soldégn Interop, teams are returning to their respective organizations to translate the week’s prototypes into production-ready code. The coming weeks will be characterized by intensive efforts to harden client implementations against the new specifications, finalize test coverage, and merge the draft Pull Requests generated at Soldégn.
As is standard practice, final decisions on key values, such as the 200 million gas limit target and the precise repricing numbers, will be made and publicly communicated during AllCoreDevs calls. These decisions are expected to be the primary focus of upcoming discussions.
The success of the Soldégn Interop is a testament to the dedication of the over one hundred core contributors who traveled to 78° North. Special commendation is due to EthPandaOps for their daily coordination and motivation of the group, and to all who worked under the perpetual daylight to achieve daily goals. The event also saw the participation of the Ethrex crew, marking their first interop. The week was exceptionally productive, and the anticipated release of a full short film promises to serve as a valuable record of this significant gathering.
