Ethereum Draft Targets Transaction Finality at Least 4 Times Faster
Key Facts
A draft proposal for Ethereum's "decoupled consensus" was submitted on 21 October 2026, with an initial goal of making transaction finality at least 4 times faster, according to the draft published in the network's improvement proposal repository. The request to add the draft remained open, with the speed target dependent on parameter selection and further analysis. Ethereum's official roadmap estimates the current time for a block to reach finality at about 15 minutes, a wait the development effort seeks to shorten. The proposal puts changes to how validators vote at the heart of that effort, separating finality voting from voting to select the chain head. In an explanation published on the Ethereum Foundation's consensus platform, Ben Edgington links the main benefits of faster finality to bridges between networks and interoperability.
Edgington explains that validators will vote independently on finality and chain-head selection, with votes transmitted and processed separately. He said validators "will cast separate votes for finality and fork choice," explaining the design underpinning decoupled consensus. Official research explanations preceding the draft proposed using a committee of 256 or 512 validators to select the chain head. That approach linked committee size to the ability to collect members' votes more quickly, as part of the search for a faster mechanism for this part of consensus. The official roadmap places these efforts within a trade-off between computing requirements, decentralization and the time needed to reach finality. The research effort thus combines a redistribution of voting tasks with consideration of the level of participation required for each task.
On practical applications, Edgington said "the main benefits of fast finality will be fast, low-cost bridges between networks and interoperability." This assessment focuses on the wait for settlement to complete when assets move between networks, and on the cost of executing those transfers. Ethereum Foundation research also examined stakeholders' finality needs, drawing on 19 interviews with individuals and institutions. The interviews were supplemented by written feedback from other participants as part of research published before the proposal was submitted. This research context brings together the views of finality users and institutions with technical work to change the voting mechanism that determines when it is reached.
Finality under the current proof-of-stake mechanism rests on a cryptoeconomic guarantee, according to Ethereum's official website. Reversing a finalized block requires burning at least 33% of all staked ether. This requirement represents the economic cost of reversing a block after it reaches that stage, and underpins the protection provided by current finality. The draft lists improved resistance to chain reorganizations among the proposed design's goals, while further analysis of safety, liveness and recovery remains necessary. Faster settlement is therefore being studied alongside an assessment of the design's ability to keep the network operating and recover from disruptions, within the broader trade-off between speed, operational requirements and decentralization.
On implementation, the Ethereum Foundation said in its protocol priorities overview that work on specifications and prototypes was under way for a subsequent upgrade, with decoupled consensus the leading candidate to headline it. Proposal development moves from research to drafting an improvement proposal, then building a prototype and testing it on a development network. This is followed by successive selection stages before reaching the mainnet, according to the sequence published by the foundation. This process places specification development and testing before activation of the change on the mainnet. In the broader roadmap, the foundation aims to make Ethereum's base layer quantum-resistant by December 2029. That goal covers the execution, consensus and data layers, extending security development across the network's core components alongside work to accelerate finality.