PropAMM Rises: Where Does the Profit Advantage Come From, and Who Bears the Cost of Deceptive Quotes?

Written by: Lioba Heimbach, Jason Milionis, Category Labs

Translated by: AididiaoJP, Foresight News

Traditional AMM (like Uniswap) liquidity providers are passive and endure adverse selection over the long term. The prices in the pool only fluctuate with trades, and once the external reference price changes, the outdated quotes of LPs can be harvested. To combat this issue, PropAMM emerged in 2024. In PropAMM, a single operator quotes with their own inventory, actively repricing, with pricing, scale, and risk management all being proprietary logic. Today, PropAMM has handled a large volume of transactions across multiple chains such as Monad.

What does this new design mean for LPs and retail investors? Is the PropAMM operator really more advantageous than passive LPs? Where do these advantages come from? Can it provide tighter quotes for retail investors? To answer these questions, we measured PropAMM activity on Solana, Base, and Monad throughout the year from September 2025 to August 2026 and decoded the closed-source logic of the largest PropAMM, Tessera, on Base.

Core Conclusions

• PropAMM is more profitable than passive LPs. Two seconds after a transaction, PropAMM earns 0.37, 1.19, and 1.69 basis points on Solana, Base, and Monad respectively, while AMM passive LPs lose 0.22, 0.62, and 2.06 basis points.
• Excess returns come from four aspects: cheap and frequent repricing, pricing based on counterparties, and arbitrage against AMM.
• PropAMM offers retail investors tighter quotes than AMM across all three chains. For example, on Monad the quotes are 1.62 to 8.60 basis points.
• Users bear the cost of deceptive quotes: after adjusting fees, traders' actual transactions often end up worse than the visible quotes.

How Does PropAMM Work?

The AMM price is determined by pool reserves, and only trades drive price changes. PropAMM, however, is priced by the operator with proprietary logic. The operator can send updates at any time to move the quotes without the need for swaps. Updates are cheap because only a few parameters need to be changed, rather than every quote. For instance, on Base, Tessera sets each tier of the order book at a fixed multiplier of a single anchor price and spread. Updating these two values can move the entire order book.

In addition to quotes, PropAMM can also charge penalty fees. Penalties could depend on the age of the quote, the source of the eating order, trading priority fees, and their relative update positions in the block.

PropAMM Landscape: Solana Leads, Base and Monad Follow

In the sample year, PropAMM on Solana averaged daily trading of 1.12 billion USD of SOL/USDC, while AMM averaged 292 million USD. Trading volumes for both types of venues are declining, with PropAMM dropping from an average of 1.82 billion USD in the first three months to 527 million USD in the last three months.

Base followed later, with its first PropAMM launching on October 30, 2025. The ETH/USDC trading volume increased from an average of 9.5 million USD in the initial three months to 29.6 million USD in the final three months, with AMM's share rising from 2.4% to 10.9%.

The first PropAMM on Monad launched in early April 2026. Since then, the daily trading volume on PropAMM for MON/USDC increased from 2.1 million USD to 3.9 million USD, with AMM's share rising from 12.5% to 45.7%.

Can PropAMM Really Outperform AMM?

To measure whether it outperforms, we look at markout, which is the profit or loss at a fixed point relative to the reference price after the transaction. Each transaction is assessed using Bybit's micro prices, weighted by notional amount. A positive value indicates LPs are making money.

Data shows AMM has varying degrees across the three chains, but the shapes are consistent. Prior to the transaction, markout is strongly positive: in the five seconds before the transaction, Solana, Base, and Monad are 5.34, 3.95, and 7.76 basis points respectively; it sharply turns negative right around the transaction. This is because AMMs only reprice after a trade occurs, and when the reference price moves, the outdated quotes of LPs get harvested.

The PropAMM curve is the opposite. AMM earns before the transaction and loses after; PropAMM loses before and earns after. For Solana and Monad, the markout for the five seconds before the transaction is -0.69 and -2.65 basis points respectively, which rapidly rise near the transaction and remain positive afterwards. Base is an exception as the markout never turns negative. Notably, the transaction does not arrive randomly, but follows reference price movements. However, PropAMM still manages to earn after transactions. This is consistent with it repricing with the reference price and not becoming the harvested party.

Result: PropAMM is more profitable than passive LPs. Two seconds after a transaction, PropAMM earns 0.37, 1.19, and 1.69 basis points on Solana, Base, and Monad respectively, while AMM passive LPs lose 0.22, 0.62, and 2.06 basis points.

Where Do Excess Returns Come From?

Cheap and Frequent Repricing

PropAMM updates quotes based on transactions, such as changing anchor prices, which is much cheaper than swaps. On Solana, median updates consume 485 to 676 computing units, while swaps require at least 16,938 units, making update costs an order of magnitude lower. On Base and Monad, updates are 5 to 15 times cheaper than swaps. All three chains default to prioritizing transactions based on computing unit fees, allowing PropAMM to gain low-cost access to early block positions.

Implementation details of Tessera: On Base, Tessera pays a base fee of 2.37 times for price updates and only 0.18 times for order book updates, meaning it only pays to secure positions when time is critical. It updates more frequently during high volatility. Within a five-minute window, the number of price updates is correlated with ETH realized volatility at 0.56.

Pricing Based on Counterparties

Among the six main PropAMMs on Solana, five provide tighter spreads for aggregator traffic compared to non-aggregators. We measure the backward spread, which is the execution of two trades in the same pool in opposite directions, of similar size, without any quote updates in between. Differences are due to PropAMM. GoonFi charges 3.59 basis points more for non-aggregator traffic, while SolFi V2 charges 1.58 basis points more. The difference for BisonFi and Tessera is only 0.10 basis points.

Implementation details of Tessera: On Base, Tessera classifies addresses into: whitelisted, penalized, and a small blacklisted group. Weighted by transaction volume, the marginal trade has a cost of 0.44 basis points before fees and 1.95 basis points after fees. Whitelisted addresses have an effective marginal cost of 0.78 basis points, while penalized addresses face 4.96 basis points.

Arbitrage Against AMM

When the reference price changes, PropAMM reprices, but AMM does not, allowing arbitrageurs to trade between the two. Before such arbitrage, the reference price has already moved a few basis points. PropAMM executes trades at the updated quotes, generally balancing these legs on Solana and Base, while making a profit of 1.55 basis points on Monad. This is similar to CEX–DEX arbitrage, except PropAMM replaces CEX and both legs execute atomically on-chain.

Implementation details of Tessera: In a single transaction, Tessera earns 1.29 basis points two seconds after the transaction, but without penalties, it would lose about 1 basis point. Some transactions may be attempts at CEX–DEX arbitrage targeting Tessera. Tessera appears to be able to identify that it is being adversely selected and charges a sufficiently high penalty to transform a single transaction from a loss into a profit, rather than losing on outdated quotes.

Deceptive Quotes

Typically, aggregators route based on quotes from the end of the last block. This creates a game: PropAMM competes to post the optimal price at the end of the block because this quote determines the routing. On Base, Tessera raises the default fee early in the block and lowers it at the end. The fee increase is concentrated in the first tenth of the block, while the decrease is concentrated in the last tenth, allowing aggregators to see a lower fee.

We compare the execution price of each trade with the quotes seen by aggregators. When Tessera updates the price in between, the probabilities of the trade improving or worsening are roughly equal; when the default fee changes, 94% of the affected trade volumes execute worse.

Implementation details of Tessera: Taking a block on Base as an example. The dashed line indicates the buy and sell quotes of Tessera without fees, while the solid line includes the default fee. D indicates a change in default fees, S indicates a swap, and P indicates a price or spread update. Before the first swap in the block, Tessera raises the default fee from 0 to 538 ppm, causing the spread including default fees to widen from 2.78 to 13.54 basis points. After the last swap, it resets fees to zero, narrowing the spread from 13.70 to 2.94 basis points. Consequently, the block ends with tighter spreads recorded by many aggregators, though each swap within the block paid a wider spread.

Do Retail Investors Benefit?

The second question: Does PropAMM provide tighter quotes for users due to stronger profitability? To isolate retail flow, we classify transactions with small reference price movements of less than 1 basis point within five seconds before and one second after the transaction as "quiet trades." Quiet trades do not carry pricing information needed by market makers and align with retail characteristics.

The answer is affirmative. In quiet trades, two seconds after the transaction, PropAMM collects 0.26, 1.26, and 1.62 basis points on Solana, Base, and Monad respectively, while AMM collects 2.59, 1.38, and 8.60 basis points. Quiet trades also constitute a higher proportion of PropAMM's trading volume: 32%, 43%, and 17% on Solana, Base, and Monad respectively, compared to 9%, 16%, and 8% for AMM. Therefore, the higher profitability of PropAMM relative to AMM LPs does not come at the expense of retail investors.

Outlook

Overall, PropAMM is beneficial for both operators and retail investors. They avoid the losses of passive LPs and provide tighter quotes for retail investors. However, users currently bear the cost of deception, as aggregators route based on the last block's end quote. If there were an on-chain router querying venues within swap transactions, this gap could be closed, routing each transaction based on the actual prices obtained. PropAMM’s shared routing interface MRC-17 is a step in this direction.