MEV-Boost

MEV-Boost is a critical piece of post-Merge Ethereum infrastructure that separates block proposal from block construction. Before a validator is selected to propose a block, they can use MEV-Boost to receive a complete, pre-built block—including its header and transactions—from a network of external block builders. This system creates a competitive auction where builders bid for the right to have their block included, with the validator typically receiving the highest bid as extra reward, known as proposer payments or MEV rewards. The validator's role is reduced to signing the block header, which is provided via a trusted relay that ensures the block is valid and the payment is guaranteed.

The architecture of MEV-Boost is built around three main actors: the proposer (the validator), the builder (a specialized entity constructing profitable blocks), and the relay (a trusted intermediary that receives blocks from builders and forwards them to proposers). This separation, often called proposer-builder separation (PBS), is designed to democratize access to MEV profits and reduce the centralizing pressure of sophisticated, in-house block-building operations. By allowing validators to easily access blocks from the most efficient builders, MEV-Boost aims to distribute MEV revenue more widely across the validator set and improve network efficiency.

For the system's security, relays play a crucial role. They perform validity checks on the blocks they receive from builders—ensuring state transitions are correct and payments are bonded—before presenting them to proposers. This prevents validators from being tricked into signing invalid blocks. The most common block-building strategy involves maximal extractable value (MEV) opportunities like arbitrage and liquidations, which builders bundle into highly profitable blocks. The builder's bid to the validator is the total value of the block's transactions and MEV, minus their own profit margin.

The widespread adoption of MEV-Boost has significantly shaped Ethereum's economics. It has become the dominant source of validator rewards beyond standard issuance and transaction fees, often doubling or tripling a validator's yield during periods of high network activity. However, it also introduces reliance on a small set of dominant relays and builders, raising concerns about centralization. The long-term Ethereum roadmap addresses this by aiming to enshrine proposer-builder separation directly into the protocol's consensus layer, making the trust assumptions and incentives more robust and decentralized than the current outsourced middleware model.

MEV-Boost is a middleware protocol that enables Ethereum validators (block proposers) to source blocks from a competitive, off-chain marketplace of specialized block builders. Instead of constructing a block locally, a validator running MEV-Boost receives a complete, pre-built block from the highest bidder in an auction. This separation of roles—proposing versus building—allows validators to capture Maximum Extractable Value (MEV) revenue more efficiently while maintaining network decentralization by preventing a single entity from dominating block production.

The core workflow involves three key actors: the relay, the builder, and the proposer (validator). Builders construct blocks, bundling user transactions and MEV opportunities, then submit their blocks and a bid to a trusted relay. The relay validates the block's correctness and forwards the highest-bid block header to the proposer via MEV-Boost software. The proposer then signs this header, committing to the block without seeing its full contents, a process secured by a commit-reveal scheme. This ensures the proposer cannot steal the builder's transaction order or MEV strategy.

Critical to its trust model, a relay acts as an intermediary that attests to a block's validity—checking for proper execution, fee compliance, and lack of censorship—before presenting it to the proposer. Proposers typically connect to multiple relays to maximize competition and reduce reliance on any single operator. The winning builder's full block is only revealed to the network after the proposer has signed the header, at which point it is published to the Ethereum peer-to-peer network for verification and inclusion in the chain.

This architecture creates a more efficient and specialized MEV supply chain. Professional builders can invest in sophisticated algorithms (e.g., for arbitrage and liquidations) and dense block packing, generating higher profits. Validators, in turn, earn a larger share of these profits through the builder's bid, which is typically the transaction priority fees and MEV revenue minus the builder's margin. This market dynamic increases the overall economic security of Ethereum by distributing MEV revenue more broadly across the validator set.

The implementation relies on Ethereum's proposer-builder separation (PBS) design philosophy. While MEV-Boost is an interim, off-chain implementation of PBS, it paves the way for a potential native, protocol-level PBS feature in future Ethereum upgrades. Its widespread adoption has fundamentally reshaped Ethereum's block production, with the vast majority of blocks being built through this outsourced marketplace, highlighting its critical role in the network's current economic and operational landscape.

The journey toward Proposer-Builder Separation (PBS) began as a direct response to the risks of Maximal Extractable Value (MEV) centralization in Ethereum's proof-of-stake design. In the initial model, a single validator was responsible for both constructing and proposing a block, concentrating both MEV extraction capability and block production power. This created centralizing pressures, as validators with sophisticated MEV strategies could outbid others for stake. PBS was conceived to separate these roles: specialized block builders compete to create the most valuable blocks, while proposers simply choose the most profitable header. This separation aims to democratize access to MEV profits and keep the validator set decentralized.

MEV-Boost served as the critical, interim off-chain implementation of PBS prior to native protocol support. It is an outsourced block building middleware that allows a consensus-layer validator (the proposer) to receive full blocks from a competitive marketplace of builders. Proposers run MEV-Boost software, which connects them to a network of relays. Builders submit their block bids to these relays, which then forward the most profitable block header to the proposer. The proposer signs this header, committing to the block without seeing its contents, and only receives the full block data just before publication. This design preserved decentralization at the proposer level while enabling a professionalized builder market to emerge.

While successful, the off-chain PBS model embodied by MEV-Boost has inherent limitations. It introduces trust assumptions in relays, which must correctly propagate blocks and censor bids. It also creates complexity for validators in managing additional software and connections. Most critically, it is voluntary; proposers can still build blocks locally, which can lead to economic inefficiency and a fragmented market. These shortcomings highlighted the need for a protocol-native solution that could enforce PBS rules for all participants, eliminate intermediaries, and provide stronger cryptographic guarantees around block withholding and censorship resistance.

Enshrined PBS refers to the potential future integration of proposer-builder separation directly into the Ethereum protocol's consensus rules. This evolution would move the auction mechanism and commitment scheme on-chain, making PBS mandatory and trust-minimized. Key research directions include builder commitments through a crList (censorship resistance list) and a two-slot proposal design. Enshrined PBS aims to permanently cap the power of any single entity, ensure credible neutrality in block building, and provide enforceable anti-censorship properties, ultimately fulfilling the original decentralization goals that motivated the separation of roles in the first place.