TechFlow Logo
Login/ Sign up
ETH Gas
Gwei
Fear
gas
On-Chain Active Market Making Tested: Earning More Than AMMs—Are Retail Investors Footing the Bill?

On-Chain Active Market Making Tested: Earning More Than AMMs—Are Retail Investors Footing the Bill?

2026.10.16
Share
The Rise of PropAMM: Where Does the Profit Advantage Come From, and Who Bears the Cost of Spoofing?

Author: Lioba Heimbach, Jason Milionis, Category Labs

Compiled by: AididiaoJP, Foresight News

Traditional AMMs (such as Uniswap) feature passive liquidity providers who continuously bear adverse selection. Pool prices only change with trades; once an external reference price shifts, LPs' stale quotes get picked off. To counter this issue, PropAMMs emerged in 2024. In a PropAMM, a single operator quotes using its own inventory, actively reprices, and employs proprietary logic for pricing, sizing, and risk management. Today, PropAMMs already handle significant trading volume across multiple chains including Monad.

What does this new design mean for LPs and retail traders? Do PropAMM operators truly have an advantage over passive LPs? Where does this advantage come from? Can they offer tighter spreads to retail traders? To answer these questions, we measured a full year of PropAMM activity on Solana, Base, and Monad from September 2025 to August 2026, and decoded the closed-source on-chain logic of Tessera, the largest PropAMM on Base.

Key Takeaways

  • PropAMMs are more profitable than passive LPs. Two seconds after execution, PropAMMs earned 0.37, 1.19, and 1.69 basis points on Solana, Base, and Monad, respectively, while passive AMM LPs lost 0.22, 0.62, and 2.06 basis points.
  • Excess returns stem from four sources: cheaper and more frequent repricing, counterpart-dependent pricing, and AMM arbitrage.
  • PropAMMs offer retail traders tighter quotes than AMMs across all three chains. For example, 1.62 basis points on Monad versus 8.60.
  • Users absorb deceptive quoting: After adjusting fees, Tessera's actual trade executions often end up worse than the displayed quotes.

How do PropAMMs work?

AMM prices are determined by pool reserves, with only trades moving the price. PropAMMs, however, are priced by the operator using proprietary logic. The operator can send updates at any time to shift quotes without needing a swap. Updates are cheap because only a few parameters need changing instead of individual quotes. For instance, on Base, Tessera sets every level of its order book as a fixed multiple of a single anchor price and spread. Updating just these two values shifts the entire order book.

Beyond quoting, PropAMMs can also charge punitive fees. Penalties can depend on quote age, aggressor source, transaction priority fees, and their relative position within the block compared to updates.

The PropAMM Landscape: Solana Leads, Base and Monad Catch Up

Throughout the sample period, daily average trading volume for SOL/USDC on PropAMMs on Solana was $1.12 billion, compared to $292 million for AMMs. Trading volumes across both venues were declining; PropAMMs dropped from a daily average of $1.82 billion in the first three months to $527 million in the last three months.

Base followed later, with its first PropAMMs launching on October 30, 2025. Its ETH/USDC trading volume grew from a daily average of $9.5 million in the initial three months to $29.6 million in the final three months, increasing its share of AMM volume from 2.4% to 10.9%.

Monad's first PropAMM launched in early April 2026. Thereafter, daily average MON/USDC trading volume on PropAMMs rose from $2.1 million to $3.9 million, lifting their share of AMM volume from 12.5% to 45.7%.

Can PropAMMs Really Outperform AMMs?

To measure outperformance, we look at markout, which is the P&L at a fixed interval post-execution relative to the reference price. Each trade is evaluated using the Bybit micro-price and weighted by notional amount. Positive values indicate LP profit.

Data shows that while the magnitude varies across the three chains, the shape is consistent for AMMs. Markouts are strongly positive pre-execution: five seconds before a trade, they stand at 5.34, 3.95, and 7.76 basis points on Solana, Base, and Monad, respectively; they turn sharply negative around execution. Because AMMs only reprice after trades occur, any movement in the reference price causes LPs' stale quotes to get picked off.

The PropAMM curve, however, is inverted. While AMMs profit pre-execution and lose post-execution, PropAMMs lose pre-execution and profit post-execution. On Solana and Monad, markouts five seconds before execution were -0.69 and -2.65 basis points, respectively; they quickly rise near execution and remain positive afterward. Base is the exception, where markouts never turned negative. Notably, executions are not random; they follow reference price movements. Yet PropAMMs still profit post-execution. This aligns with their ability to reprice alongside the reference price rather than becoming victims of picking.

Result: PropAMMs are more profitable than passive LPs. Two seconds after execution, PropAMMs earned 0.37, 1.19, and 1.69 basis points on Solana, Base, and Monad, respectively, while passive AMM LPs lost 0.22, 0.62, and 2.06 basis points.

Where Does the Excess Return Come From?

Cheaper and More Frequent Repricing

PropAMMs adjust quotes via update transactions, such as modifying anchor prices, which are far cheaper than swaps. On Solana, median updates consume between 485 and 676 computational units, compared to at least 16,938 for a swap—updating costs more than an order of magnitude less. On Base and Monad, updates are 5 to 15 times cheaper than swaps. All three chains prioritize transactions based on compute-unit fees by default, allowing PropAMMs to secure front-of-block positions at low cost.

Tessera implementation details: On Base, Tessera pays 2.37 times the base fee for price updates but only 0.18 times for order book updates, meaning it only pays to grab positioning when timing is critical. It updates more frequently during high volatility. Over a five-minute window, the correlation between its number of price updates and ETH realized volatility is 0.56.

Counterparty-Dependent Pricing

Of six major PropAMMs on Solana, five offer tighter spreads for aggregator traffic than for non-aggregator traffic. We measure round-trip spreads, defined as two opposing-direction, similarly sized executions in the same pool with no quote updates in between. The gap varies by PropAMM. GoonFi charges 3.59 basis points more for non-aggregator traffic, and SolFi V2 charges 1.58 basis points more. The difference for BisonFi and Tessera is merely 0.10 basis points.

Tessera implementation details: On Base, Tessera categorizes addresses into whitelisted, penalized, and a small direct blacklist category. Weighted by trading volume, the marginal cost per trade is 0.44 basis points pre-fee and 1.95 basis points post-fee. Whitelisted addresses incur an effective marginal cost of 0.78 basis points, while penalized addresses face 4.96 basis points.

Arbitraging AMMs

When the reference price moves, PropAMMs reprice while AMMs do not, prompting arbitrageurs to trade between them. Prior to such arb legs, the reference price has already shifted by several basis points. Executing against updated PropAMM quotes yields roughly break-even legs on Solana and Base, and earns 1.55 basis points on Monad. This resembles CEX–DEX arb, except the PropAMM replaces the CEX, and both legs execute atomically on-chain.

Tessera implementation details: On a per-trade basis, Tessera profits 1.29 basis points two seconds after execution; without penalties, it would lose approximately 1 basis point. Some executions may be CEX–DEX arb attempts targeting Tessera. Tessera appears capable of identifying when it is being adversely selected and levies sufficiently high penalties to flip losing trades into profitable ones, rather than suffering from stale quotes.

Deceptive Quoting

Typically, aggregators route based on the quotes at the end of the previous block. This creates a race: PropAMMs compete to post the best prices at block ends since those determine routing. On Base, Tessera increases its default fee early in the block and decreases it toward the end. Fee hikes concentrate in the first tenth of the block, while cuts concentrate in the last tenth, ensuring aggregators read lower fees.

We compare each trade's execution price against the quotes seen by aggregators. When Tessera updates prices mid-block, trades are roughly equally likely to improve or worsen; when default fees change, 94% of the affected volume executes worse.

Tessera implementation details: Taking one Base block as an example. The dashed line represents Tessera's bid/ask without fees, while the solid line includes the default fee. D denotes default fee changes, S denotes swaps, and P denotes price or spread updates. Before the block's first swap, Tessera raises the default fee from 0 to 538 ppm, expanding the fee-inclusive spread from 2.78 to 13.54 basis points. After the final swap, it resets the fee to zero, narrowing the spread from 13.70 to 2.94 basis points. The block thus closes with tight spreads read by many aggregators, even though every swap within the block executed at a wider spread.

Did Retail Traders Benefit?

Second question: Do PropAMMs offer users tighter quotes precisely because they are more profitable? To isolate retail flow, we define "quiet executions" as trades where the reference price moved less than 1 basis point from five seconds pre-execution to one second post-execution. Quiet executions carry no pricing information that market makers need to react to, aligning with retail characteristics.

The answer is yes. For quiet executions, PropAMMs charged 0.26, 1.26, and 1.62 basis points two seconds post-execution on Solana, Base, and Monad, respectively, while AMMs charged 2.59, 1.38, and 8.60 basis points. Quiet executions also make up a larger share of PropAMM volume: 32%, 43%, and 17% on Solana, Base, and Monad, versus 9%, 16%, and 8% for AMMs. Therefore, PropAMMs' higher profitability relative to AMM LPs does not come at the expense of retail traders.

Outlook

Overall, PropAMMs benefit both operators and retail traders. They avoid the losses suffered by passive LPs while providing tighter quotes to retail. However, users currently absorb spoofing costs because aggregators route based on quotes at the end of the previous block. If an on-chain router could query various venues within a swap transaction itself, it would close this loophole and route each trade based on the actual executed price. The PropAMM shared routing interface, MRC-17, represents a step in this direction.