Yes. A Category Labs study measured Solana and Base from September 2025 to August 2026, and Monad from April 2026. Two seconds after a fill, propAMMs earned 0.37, 1.19 and 1.69 bps, while passive AMM liquidity providers lost 0.22, 0.62 and 2.06 bps. PropAMMs also quoted retail tighter, at 1.62 against 8.60 bps on Monad.
Ozan Solmaz, Lioba Heimbach and Jason Milionis wrote the paper (September 2026) and a Category Labs summary (14 October 2026). They measure markouts: the profit or loss on each fill against the Bybit microprice at a fixed time after the trade, weighted by size. A positive markout means the liquidity provider gained.
How does a propAMM work?
A single operator quotes from its own inventory and moves the price whenever it wants. On a classic AMM such as Uniswap, the pool's reserves set the price, so the price changes only when someone trades. When the outside price moves, the pool keeps quoting the old price until an arbitrageur trades against it. The liquidity providers absorb that loss.
A propAMM operator sends a small update transaction instead. Tessera, the largest propAMM on Base, places each level of its book at a fixed multiplier from one anchor price and spread, so changing those two values moves the whole book (X article). The operator can also add penalties on top of the quote, based on the quote's age, who the taker is, the priority fee paid and where the trade sits in the block.
The markout curves show the result. On Monad, AMM liquidity providers sit 7.76 bps in the money five seconds before a fill and lose after it. PropAMMs start 2.65 bps out of the money and earn after it (Category Labs, Sep 2025 to Aug 2026).
Where does the propAMM edge come from?
The authors trace it to four sources. All figures below come from the X article and cover September 2025 to August 2026.
- Cheap, frequent repricing. On Solana, a median update uses 485 to 676 compute units against at least 16,938 for a swap. On Base and Monad, an update is 5 to 15 times cheaper than a swap. All three chains order transactions by priority fee per unit of compute by default, so a cheap update buys an early block position for little.
- Pricing by counterparty. Five of the six largest Solana propAMMs give aggregator flow tighter spreads than other flow. GoonFi charges non-aggregator flow 3.59 bps more and SolFi V2 1.58 bps more. On Base, Tessera sorts addresses into classes: whitelisted addresses pay an effective marginal cost of 0.78 bps, penalised ones pay 4.96 bps.
- Arbitrage against AMMs. When the outside price moves, the propAMM reprices and the AMM does not, so arbitrageurs trade between them. The propAMM roughly breaks even on these legs on Solana and Base and earns 1.55 bps on Monad.
- Spoofing. Aggregators route on the quote from the end of the previous block. Tessera raises its default fee early in the block and lowers it at the end, so the aggregator reads the low fee and the trade pays the high one.
Penalties carry a lot of the weight. Tessera earns 1.29 bps on single fills two seconds after the fill. Without its penalties it would have lost about 1 bp.
What does the data show on Solana, Base and Monad?
PropAMMs carry most of the volume on Solana and are gaining share on Base and Monad. Volumes are daily averages for the named pair, per the X article.
| Solana (SOL/USDC) | Base (ETH/USDC) | Monad (MON/USDC) | |
|---|---|---|---|
| Period | Sep 2025 to Aug 2026 | First propAMMs from 30 Oct 2025 to Aug 2026 | First propAMM from early Apr 2026 to Aug 2026 |
| PropAMM volume per day | $1.12B average, against $292M on AMMs | $9.5M in first three months, $29.6M in last three | $2.1M rising to $3.9M |
| PropAMM share against AMMs | Majority of volume | 2.4% to 10.9% | 12.5% to 45.7% |
| PropAMM markout, 2s after fill | +0.37 bps | +1.19 bps | +1.69 bps |
| AMM LP markout, 2s after fill | -0.22 bps | -0.62 bps | -2.06 bps |
| Retail cost, propAMM vs AMM | 0.26 vs 2.59 bps | 1.26 vs 1.38 bps | 1.62 vs 8.60 bps |
| Quiet fills share of volume, propAMM vs AMM | 32% vs 9% | 43% vs 16% | 17% vs 8% |
Solana volume fell on both venue types over the year. PropAMM volume there dropped from $1.82B per day in the first three months to $527M in the last three.
The retail rows use "quiet fills", where the reference price moves less than 1 bp from five seconds before the fill to one second after. These fills carry no information the market maker has to price against, so they stand in for retail. The gap is widest on Monad and smallest on Base.
Who pays for the propAMM edge?
Users pay for the spoofing. The authors compared each trade's execution with the quote the aggregator saw. When Tessera only changed its price between the two, trades were about as likely to do better as worse. When the default fee had changed, 94% of the affected volume did worse than quoted (X article, Sep 2025 to Aug 2026).
One example block on Base shows the pattern. Tessera raised its default fee from 0 to 538 ppm before the first swap, widening its spread after fees from 2.78 to 13.54 bps. After the last swap it reset the fee to zero and the spread narrowed from 13.70 to 2.94 bps. Aggregators read the tight spread at the end of the block. Every swap inside the block paid the wide one.
The authors propose a fix: an on-chain router that queries each venue inside the swap transaction, so each trade routes on the price it actually gets. They point to MRC-17, a shared routing interface for propAMMs, as a step toward it.
What does this mean for teams building or integrating on-chain liquidity?
PropAMMs now set many of the prices retail users see, and aggregator routing decides which venue gets each trade. Each type of team has a different job.
| Team | What the data says | What to do next |
|---|---|---|
| Market makers | Edge comes from cheap updates, block position and counterparty pricing | Price update cost and transaction ordering on each target chain before you commit inventory |
| DEX aggregators | End-of-block quotes let a venue show one price and fill at another | Measure delivered price against quoted price per venue, and evaluate in-transaction routing such as MRC-17 |
| Wallets | Retail flow is the cheapest flow to fill, yet users can pay more than quoted | Choose routing partners on delivered price, and ask them for the quote-to-fill gap by venue |
| Chains | Monad propAMM share rose from 12.5% to 45.7% from April to August 2026 | Court propAMM operators early, since cheap updates and priority ordering decide where they deploy |
For revenue leaders, the counterparty finding matters most. Five of the six largest Solana propAMMs price aggregator flow tighter, so aggregators are the sales channel for a propAMM. Map every aggregator and wallet router on its target chain, name the integration owner at each one, and track weekly which ones route to you. Our notes on how blockchain infrastructure protocols grow market share cover that integration sequence. Our team worked on Monad infrastructure at FastLane, where shMonad runs liquid staking on Monad.
To map the aggregators, wallets and market makers in your market and the state of your integration pipeline, start with The Diagnostic at $2,500.