Revolutionary auction design with multidimensional pricing cuts gas costs 30% while guaranteeing validator income. Solve MEV extraction inefficiencies.
Traditional PBS (Proposer-Builder Separation) designs often promote "proposer dumbness," where passive proposers (i.e. validators) simply accept the highest bid for their entire block space. XGA challenges this constraint by enabling flexible commitment (re: allocation) mechanisms.1
Multidimensional fees have always been coached in terms of “price”. Calculating transaction fees through a single, joint unit of account fixes the relative prices of all resources.
What resources could be multidimensionally priced?
We can start with the basics:
EVM execution
Tx calldata
Witness data
Storage size growth
source: Multidimensional EIP 1559 , Vitalik Buterin
The fixed relative prices inhibit granular price discovery and, ultimately, reduce the throughput of the blockchain. This is an accurate assessment, and we are going to completely ignore it by focusing on quantity , instead of price.
Instead of approaching Multidimensional fees as a “re-pricing” problem, we instead reframe it as a “quantity” allocation problem. This is how the Volumetric Block Auction works: it can be thought of two auctions as:
Determines who gets in
Determines who gets specific contested resources
Instead of adjusting the block size, it creates two markets for parts of the total block size. XGA, the forward call option market, is the “Determines who gets in” auction.
The call option bid is only charged at delivery: when you exercise the option by submitting a transaction bundle to the XGA relay.
It determines a threshold ratio µ such that all transactions and bundles with a bid-to-size ratio at least µ can fit into the block capacity.
µ.Specifically, for an included bundle corresponding to the transaction t with the highest bid bt, the searcher pays the maximum of the threshold price based on the bundle size (µ × s't), and the second-highest bundle bid for that same transaction (bt'). If there's no second bid, bt' it is 0.
µis non-decreasing with the number of bids
Transactions and bundles are included if their bid-to-size ratio meets or exceeds this threshold. This is a fundamental difference in how the quantity of block space is determined and allocated, moving beyond a simple price threshold.2
How can we enforce this commitment to the auction outcome?
There are two models: Optimistic and Pessimistic.
The critical distinction lies in scenario #4, where proposers make commitments but then deviate from them. This represents an invalid state transition, and is currently only punishable (“slashing after the fact”), i.e. optimistic enforceable. Pessimistic would mean that it would be impossible, meaning this couldn’t happen as it is enforced in protocol (i.e. Ethereum consensus enforced).
optimistic (validators can violate but get slashed)
pessimistic (violations make blocks inherently invalid)
In the optimistic systems, proposers can create invalid blocks that violate their commitments, with the expectation that attesters will slash them afterward. However, this approach maintains misaligned incentives because the damage may already be done by the time punishment occurs.
Think of commitments as a type of governance cost.3
By leveraging re-staking, we can enforce block production commitments using optimistic (validators can violate but get slashed) commitments. Moving to pessimistic models (violations make blocks inherently invalid) isn’t necessary: commitment violations are highly unlikely to be multi block.
From a consumer's perspective, all costs can be seen as transaction costs, including the inconvenience of buying in addition to the price. This is why searchers pay the fee directly, in exchange for direct access to block space bypassing builders entirely.
High‐specificity pushes participants to demand longer-horizon payoffs (rev-share, exclusive slots, governance tokens). Accessing the XGA Auction does not directly require staking a token, instead the captive insurance pool sponsors (i.e. whitelists) searchers whom have exclusive access. This underwrites (part) of the risk.
Captive Insurance is the "Profit Protection Protocol (PPP)", it is a vault in which single-sided staking of FOLD token provides a centralized counterparty for underwriting the risk (i.e. lower yield).4
Token reward mechanics govern capital flows. This governs capital. Multipliers are rewards which increase when participants engage across multiple roles
The cost of placing a bid that was eventually executed is platform fees
Any attempt to introduce a new blockspace market faces the classic chicken-and-egg dilemma: validators won't commit supply without user demand, and users won't commit demand without guaranteed supply and liquidity. Launching a full-scale market from day one, expecting widespread adoption, is an unrealistic and dangerous "boil the ocean" strategy that has historically led to failure.
The incentives program is fundamentally about avoiding the "Boil the Ocean" trap.5 This is a critical antipattern: Any strategy requiring near-universal adoption (e.g., 50% of validators) before offering meaningful value must be rejected. This is why XGA is backwards compatible with MEV Boost: the only change needed is connecting to the XGA Relay from a validator's perspective. In order to earn the extra incentives validators must be running the XGA Commit Boost Module or use Vouch (Vouch already supports XGA, Commit Boost is forthcoming). You can use MEV Boost and earn additional yield, but it won't be as much.
Adopting XGA is just adding a new relay endpoint in MEV Boost. That’s it.
By starting with micro-commitments, guaranteeing rewards for supply, and de-risking demand, we create a viral loop of adoption:
We make participation easy for everyone:
Instead of full blocks, validators “pledge” to reserve a portion of their blockspace – "Commitment Slots" – for the separate, continuous auction (XGA). These commitment-slots are always available for the MEV Boost auction regardless.
By construction, Validators are always better off because the auction guarantees at least the same results as the normal Flashbots Auction.
Validators who commit to providing a minimum number of micro-slots per block or epoch receive a guaranteed "Participation Bounty" (e.g., 0.005 ETH per committed block) for the first X months, regardless of if the commitment-slot sells. They also receive the full proceeds from any successful commitment-slot auction.
Extra, guaranteed income just for optioning a portion of blockspace for a new market via XGA6
Searchers who bid on auctioned blockspace receive a "Guaranteed Inclusion" commitment. Searchers only pay their bid at the time of execution: This eliminates the biggest risk for auction participants: paying for bids and not receiving it.
If their transaction is not included in a winning bid execution despite submitting a transaction bundle (meaning they are exercising their option, i.e. they submitted a transaction for usage in the space they bought), a substantial portion (e.g., 50-90%) of their bid is automatically rebated by the protocol. This is the auction settling the contract by cash instead of delivery.
"I can pay for guaranteed inclusion for my critical transaction/bundles without worrying about wasting money if the system isn't liquid yet."
This market dysfunction has created a classic coordination problem. Validators hesitate to commit blockspace to new pricing mechanisms without assured demand, while users avoid bidding in untested markets without guaranteed supply. The result is economic deadlock that leaves billions in value creation on the table.
Source: Protocol Research Call 1
The question facing Ethereum stakeholders is not whether more efficient blockspace markets will emerge, but whether they will emerge on Ethereum or migrate to competing networks offering superior market structure. The window for maintaining competitive advantage through infrastructure innovation remains open, but it will not remain so indefinitely.
Validator not dumb: better markets → higher validator revenue
More competitive builder market → higher validator revenue. Weak builder market → smaller revenue. ↩
These fees are just one source of revenue for the Auction. ↩
A joke concerning marriage was intended here, oh well. If you found this funny: Share ↩
By construction, Validators are always better off because the auction guarantees at least the same results as the normal Flashbots Auction. Thus, it can be said that there isn't any risk of making lower yield. This can be seen as the risk free rate of return for those who are staking the FOLD token in the PPP Vault. ↩
Requiring a massive, coordinated adoption from a huge number of users or participants before we can deliver any value is problematic. ↩
Commitment through the call option. ↩