A new academic preprint published on Sept. 29 demonstrates that proprietary automated market makers (propAMMs) on Solana (SOL) achieved an average reference-relative execution cost of 0.26 basis points for quiet-market SOL/USDC fills, compared to 2.59 basis points for conventional public AMMs.
Authored by researchers affiliated with ETH Zurich and Category Labs, the study evaluated trading data from Sept. 1, 2025, through Aug. 31, 2026, along with shorter observation windows on Base and Monad. Trade fills were weighted by notional size against Bybit's size-weighted top-of-book USDT microprice, adjusted using the exchange's USDC/USDT midpoint.
Execution Efficiency vs. Liquidity Provider Returns
Across the Solana dataset, the paper recorded two-second gross maker markouts of +0.37 basis points for operator-controlled propAMMs, compared to -0.22 basis points for public AMMs. The quiet-flow execution proxy measured trades where reference prices moved less than 1 basis point from five seconds prior to one second following execution.
While lower swap costs benefit traders seeking optimal token outputs, passive liquidity providers (LPs) in public pools remain exposed to arbitrageurs picking off stale quotes when external markets move first. Loss-versus-rebalancing models highlight that arbitrage drag reduces overall LP returns alongside inventory shifts, transaction expenses, and hedging costs.
- 0.26 basis points: Execution cost proxy for quiet SOL/USDC trades on propAMMs versus 2.59 basis points on public pools.
- +0.37 vs. -0.22 basis points: Two-second gross maker markout differential favoring institutional propAMMs over passive AMMs.
- Sept. 1, 2025 – Aug. 31, 2026: Core sampling timeframe covering Solana, alongside metrics from Base and Monad.
Quote Reliability and Routing Protections
To manage adverse selection risk, professional market makers adjust quoting behavior dynamically. An April account from Jump Crypto detailed how its BisonFi propAMM tailors available depth and pricing based on quote freshness, inventory levels, and incoming flow quality. Aggregators like Jupiter utilize dedicated signers to identify non-toxic retail orders originating from frontend interfaces.
However, price display discrepancies persist across decentralized venues. On Base, metrics for Tessera revealed execution averaged 1.08 basis points worse by trade count and 0.56 basis points worse by volume relative to reconstructed previous-block-end quotes. This block-timed fee phenomenon aligns with findings from a March 20 report by routing protocol 0x, which noted price degradation between order selection and settlement across unnamed operators. Aggregators such as Jupiter continue to implement quote-to-execution parity tests and router competition to penalize underperforming liquidity sources across DeFi protocols and stablecoin trading networks.
Why It Matters
This research highlights a growing structural divide in decentralized finance between institutional market makers and retail liquidity providers. While propAMMs deliver vastly superior execution prices for ordinary swappers, they do so by actively shielding themselves from toxic flow—a protection that passive public pools lack. As DEX routers refine real-time execution comparisons, capital is likely to concentrate further into professionally managed pools, forcing public AMMs to redesign fee structures or risk complete yield decay.
