$61,630-3.47%
BTC7D TREND
$1,746-3.38%
ETH7D TREND
$72.12-3.25%
SOL7D TREND
$570.47-3.38%
BNB7D TREND
DeFi Data →
Liquidity
Liquidity

Anatomy of a Fill: How PropAMMs Work on Ethereum

News

The practical difference for a DeFi trader is that a PropAMM fill is a firm quote, not a reserve ratio. The maker controls what price is executable at any moment, can widen spreads when inventory skews, and can withdraw quotes entirely.

·7 min readMakeDefilibanpreferred onGoogle

This article synthesizes reporting and analysis from azfuller, LambdaClass, and POL Finance. Proprietary AMMs on Ethereum have crossed $3.35 billion in all-time volume across 277,834 swaps, according to independent researcher azfuller, with two venues alone trading nearly $460 million in a single week in September 2026 and capturing a peak 11% share of Ethereum DEX volume. A detailed post-mortem published on ethresear.ch by researchers santidipaolo, jpinasco, and agustinpolfinance of POL Finance on November 19, 2026 dissects exactly how those fills happen and what they cost the market maker in unhedged exposure during the blind window between quote signature and block arrival.

What a PropAMM Is and Why Its Liquidity Behaves Differently

A PropAMM, or proprietary automated market maker, is an on-chain liquidity pool whose quoted prices are continuously overwritten by a single market maker, potentially hundreds of times per block. Unlike a constant-product x*y=k pool where reserves are passive and any taker can sweep the curve, a PropAMM operator writes executable prices from inventory, real-time market data, and risk limits, then publishes those quotes to supported block builders. For related coverage, see NullMask: Unverified Documentation Summary of Season One.

The practical difference for a DeFi trader is that a PropAMM fill is a firm quote, not a reserve ratio. The maker controls what price is executable at any moment, can widen spreads when inventory skews, and can withdraw quotes entirely. Capital that would sit dormant across the full Uniswap v3 range is instead concentrated where demand exists. Azfuller’s data memo discusses capital turnover at Fermi and comparable Uniswap v3 pools, but does not specify the observation date, measurement window, pool-selection method, liquidity definition, or turnover calculation for the comparison. Context on this theme: DeFi Tax-Loss Harvesting Before Year-End 2026: What US Rules Say Today, and What to Check First.

According to LambdaClass’s protocol blog, Titan Builder published a PropAMM API in May 2026. This article is limited to that documented API description and does not establish an Ethereum protocol or client version, contract deployment block, block timestamp, or effective block for the API state. Quasar, BuilderNet, and Bombora subsequently adopted the same API. Those four builders now carry all PropAMM venues, and the ordering guarantee that makes the model work, that the maker’s most recent price update lands before the taker’s transaction in the same block, is enforced by their cooperation rather than by protocol rules. As the ethresear.ch post states directly: “That ordering is a promise by four companies, enforced by nothing in the protocol.” More on this in Pendle Launches USDx and sUSDx Markets From Axis Foundation.

TLDR KEYPOINTS

  • PropAMMs rewrite on-chain prices hundreds of times per block from a single maker’s inventory.
  • The maker operates blind for a median 1.23 seconds per fill, the gap between signing a quote and seeing the executed block; 16% of fills during this window saw the reference price move beyond the quoted half-spread.
  • All ordering guarantees currently rest on builder promises from four companies; ERC-8324 has been proposed to formalize this at the protocol level but is not yet live.

Kubi Mensah of Gattaca, Titan Builder’s parent company, described the structural advantage of the model: “The beautiful thing about prop AMMs [is] that they are on the ledger, and then anyone that interacts with the chain…can discover best prices,” as quoted in azfuller’s memo. LambdaClass identifies a PropAMMRouter contract at address 0x4DdF368080CD7946db5b459aD591c350158175e1 that routes across all PropAMM venues with a Uniswap v3 fallback and takes zero fees. This kind of intent-routing infrastructure, similar in spirit to what CoW Swap built for MEV-protected intent trading, is becoming a default aggregation layer for execution-quality-conscious traders.

Anatomy of a Fill: From Quote to Ethereum Settlement

Quote Formation

The maker runs an off-chain pricing engine that continuously calculates bid and ask quotes based on current inventory, a reference price feed such as Binance’s perpetuals mid, and a target spread sufficient to cover adverse selection risk over the blind window. The resulting quotes are signed off-chain and submitted to supported block builders, where they can be overwritten by a fresher signature before the block is sealed. The ethresear.ch post by POL Finance measured ETH/USDC spreads across live venues: Metric quotes a full spread of 0.86 basis points, Fermi at 2.34 bps, and bopAMM at 4.20 bps.

Order Acceptance and Risk Checks

To illustrate the lifecycle, consider a hypothetical USDC-to-ETH swap routed through the PropAMMRouter. The router queries each venue for a live quote, selects the best executable price, and submits a transaction that references the maker’s current signed price. The builder places the maker’s most recent price-update transaction immediately before the taker’s swap in the block ordering, locking in the quoted rate. No separate on-chain oracle or pool reserve calculation is needed; the fill price is whatever the maker’s last signed message said.

Partial fills, quote expiry, and reverts are all possible outcomes. If the maker’s latest signed quote has already expired or been superseded by a wider spread update, the transaction reverts and the user falls back to the Uniswap v3 route that the PropAMMRouter includes as a safety net. Fees appear only implicitly in the spread itself; there is no explicit fee tier the way Uniswap v3 has 0.05%, 0.30%, and 1.00% pools.

Onchain Settlement and the Blind Window

Settlement is fully on-chain: assets transfer atomically when the block is accepted, and the fill is verifiable by any observer. The maker’s risk problem is what happens between signing the quote and seeing that block. The POL Finance analysis measured this gap directly: the maker learns a fill occurred only when the block propagates to its own node, with a median latency of 1.23 seconds and a 90th-percentile latency of at least 2.18 seconds. During that blind window, the maker cannot hedge the position it has just been assigned.

The researchers quantified the cost of this exposure. In 16% of measured trades, the Binance reference price moved beyond the 0.90 bps half-spread during the maker’s blind window, meaning the fill was executed at a price the market had already moved through. That is the per-trade adverse selection rate that the spread must compensate for over time. Titan Builder, which built 44% of all Ethereum blocks in the measured period but carried 69% of PropAMM fills, effectively concentrates this flow, giving the maker more predictable execution timing when Titan is the winning builder.

The Trade-Offs: Tighter Fills, Managed Inventory, and Execution Risk

For traders, the primary benefit of a PropAMM fill is price certainty within the quoted size. Because the maker concentrates capital near current market price rather than spreading it across a full curve, price impact for small to mid-size swaps is lower than on a comparable passive pool. According to azfuller’s data memo, Fermi’s $280 million and Metric’s $178 million in weekly volume for the week of September 14 to 20, 2026, together representing approximately 7.6% of Uniswap’s $6.0 billion weekly volume, suggest real taker demand for this execution quality even at an early stage of adoption.

The limitations are structural. Maker capacity is finite and inventory-constrained; a PropAMM can and does widen quotes or go offline when the maker’s risk limits are breached. Quote withdrawal means a taker may see an attractive price in a simulation that is no longer available at execution time. This is behaviorally different from a passive pool, which remains executable at some price regardless of market conditions.

Dependency on builder cooperation introduces a protocol-layer risk that passive AMMs do not carry. If a builder decides not to honor the ordering guarantee, the maker’s price update and the taker’s swap can land in the wrong order, resulting in a fill at a stale price. ERC-8324 has been proposed to provide protocol-level enforcement of this ordering, according to the POL Finance ethresear.ch post, but that standard is not yet live. Until it is, PropAMM participants are trusting a four-company coordination agreement, not a smart contract invariant.

Before routing significant volume through a PropAMM, traders and integrators should verify: whether the venue publishes its quote-expiry and fallback logic; what the maker’s stated inventory limits are per token pair; which builders carry the venue and what percentage of recent blocks they built; and whether the smart contract has been audited and what upgrade mechanisms exist. The spread data, 0.86 bps at Metric versus 4.20 bps at bopAMM, is itself a signal about how each maker has calibrated adverse selection risk versus capital deployment, and comparing spreads across venues over time is the most direct measure of competitive execution quality available on-chain.

Editorial credits

Writer · Defiliban

Editor · Defiliban
Monitors published articles around the clock and can correct or withdraw them.

Checked by Defiliban QC Team before publishing on November 20, 2026 for headline accuracy, original value, factual consistency, metadata, commercial disclosure and authorship claims: our AI review system checks every article before it goes live, and our editor monitors it after publication. Report an error.

Defiliban uses AI for research, drafting and review. Our editorial team is responsible for every article and corrects or withdraws articles when needed. How we review.

Disclaimer: This article is for informational purposes only and does not constitute financial or investment advice. Cryptocurrency and digital asset markets carry significant risk. Always do your own research before making decisions.

Defiliban · Oliver Benjamin

This article was produced with AI assistance and reviewed by a human editor.

Oliver Benjamin

Written by

Oliver Benjamin

DeFi Research Writer

Oliver Benjamin is a DeFi research writer at DefiLiban, specializing in decentralized finance systems, token economics, liquidity design, and blockchain market structure. His work focuses on analyzing how DeFi protocols function beneath surface-level narratives, with particular attention to incentives, on-chain behavior, yield systems, and emerging crypto infrastructure.


Briar Ellington

Edited by

Briar Ellington

Editor

Briar Ellington is an editor at Defiliban. Briar reviews news and analysis for sourcing, accuracy and attribution.