Gate Research Institute: Prop AMM Reshapes Solana's Liquidity Landscape

Summary

• Prop AMM accounts for 20%-40% of Solana’s weekly trading volume, with over 80% in SOL-Stablecoins.
• The proprietary capital model of Prop AMM eliminates impermanent loss (IL), converting it into hedgeable inventory risk.
• Profits are not dependent on token subsidies; Prop AMM earns spreads, with profit = total spread earned from market making - market value fluctuations of inventory assets - operating costs.
• Solana is evolving towards an on-chain CEX model, with Prop AMM playing the role of a core market maker.

I. The Three Iterations of AMM Evolution

One of the core innovations of DeFi is the Automated Market Maker (AMM), which replaces traditional order book models with mathematical formulas. However, the development of AMM has not been instantaneous but has gone through three key iterations, each aimed at solving the capital efficiency issues of the previous generation.

1.1 First Generation: Standard CPMM (Constant Product Market Maker)

The first-generation AMM, exemplified by Uniswap V2 and Raydium (Legacy), ushered in the DeFi era with its simplicity. Its inherent flaw is the fixed nature of the constant product formula x * y = k, where the product of the two assets’ quantities in the pool remains constant.

This type of AMM features passive market making. Liquidity is theoretically distributed uniformly across all price ranges from 0 to ∞, ensuring liquidity at any price point. The cost is that the depth distribution is very sparse and capital efficiency is extremely low; most of the assets are located far from the current trading price, in a “sleeping” state, unable to be effectively utilized. For example, in a SOL/USDC pool with SOL priced at 150 USDC, the formula indicates that a large portion of funds are allocated at SOL prices of 1 USDC or 1,000 USDC, which are highly unlikely to be traded. This is akin to a convenience store with shelves stocked with drinks priced from 1 to 1,000 yuan, but customers only buy one or two types, leaving most goods unsold.

Meanwhile, liquidity providers (LPs) bear high impermanent loss (IL) risks, and trading fees often do not cover IL. They mainly rely on token rewards to offset losses caused by asset price volatility.

1.2 Second Generation: Concentrated Liquidity CLMM

To improve capital efficiency, the second-generation AMM introduced the concept of Concentrated Liquidity (CLMM), exemplified by Uniswap V3. It allows LPs to actively deploy funds within specific custom price ranges (e.g., only providing liquidity when ETH price is between 2,900 and 3,000 USDC), thereby offering much higher liquidity depth within the target range than CPMM. This innovation is a significant leap, with Uniswap V3 being regarded as a milestone akin to the iPhone moment in DeFi, becoming the mainstream paradigm for DeFi liquidity.

However, CLMM shifts management responsibility and complexity entirely to LPs. LPs need to continuously predict price movements and manually adjust their ranges, similar to professional market makers. If the price moves outside the set range, LP positions are fully converted into a single asset, effectively exiting market making until the price returns or the range is manually reset. This creates a fragmented user experience: active managers enjoy high yields, while ordinary retail investors face management burdens or higher risks and lower returns due to improper range settings. While improving capital efficiency, CLMM also deepens the gap between professional and amateur participants.

1.3 Third Generation: Active Management and Discrete Liquidity (exemplified by Solana)

In the previous two generations, innovation was mainly driven by Ethereum and its leading DEX Uniswap. In the current third-generation AMM revolution, Solana, with its high TPS, low transaction costs, and active on-chain activity, has become an ideal testing ground. This generation pushes liquidity management to the extreme, supporting high-frequency, millisecond-level price adjustments, achieving efficiency comparable to CEXs.

Based on Solana’s high TPS, low gas fees, and multi-transaction volume environment, the development of Solana AMMs branches into two:

(1) Branch A: Public Active DLMM, represented by Meteora Discretizes liquidity into a series of tightly connected “price bins,” characterized by transparency, permissionless access, and retail participation through algorithmic strategies within the protocol. From a technical evolution perspective, DLMM is a more elegant, user-friendly implementation and product extension of CLMM on high-performance blockchains.

(2) Branch B: Private Institutional Type (Prop AMM / Dark Pool), represented by Humidifi, Tessera V In Prop AMM, LPs are no longer retail investors but professional market makers holding large positions. It features a fully black-box approach, not relying on on-chain basic curves like x * y = k for pricing, but instead depends on external oracles and internal algorithms, mapping CEX matching capabilities and inventory management logic directly on-chain, pursuing the highest quality of trade execution.

Overall, the evolution of AMM is shifting from providing sparse liquidity everywhere to providing deep liquidity where it is most needed. Prop AMM represents another milestone in this efficiency race, driven by institutional capital offering the most professional on-chain market making services.

II. Definition and Mechanism of Prop AMM (Dark Pool)

The emergence of Prop AMM marks a key turning point in on-chain market making strategies, shifting from purely mathematical models to professional trading logic. Essentially, it is the direct on-chain deployment of high-performance market makers (MM), leveraging Solana’s low latency environment to achieve high-frequency, low-slippage trading.

2.1 Definition and Boundaries

Prop AMM refers to an on-chain trading venue fully controlled by professional market makers, with pricing algorithms that are not publicly disclosed (or partially computed off-chain), usually without a front-end interface. Due to its opaque internal operations, it is often called Dark AMM or Private Market Makers. Mainstream Solana Prop AMMs include HumidiFi, Tessera V, GoonFi, ZeroFi, and SolFi.

Figure 1: Solana Prop AMMs Trading Volume

2.2 Why “Prop” (Proprietary)?

“Proprietary” is key to understanding how these AMMs operate:

(1) Proprietary Funds: The liquidity of these AMMs is 100% from the project or its behind-the-scenes professional market makers (or the project itself is a market maker). They do not accept public deposits, fundamentally overturning the traditional DeFi reliance on retail liquidity.

(2) Proprietary Strategies: Trading strategies (how to quote, when to withdraw, how to hedge) are core business secrets of market makers. These strategies are not publicly written into smart contracts for review but are computed off-chain, with only final trading instructions sent on-chain for settlement, similar to proprietary trading desks in traditional finance.

2.3 Technical and Operational Mechanisms

Prop AMM can achieve CEX-like market making performance on-chain by relying on a sophisticated combination of off-chain high-performance computing and on-chain prioritized execution.

2.3.1 “Oracle + Solver” Hybrid Pricing Model: Off-chain Calculation, On-chain Execution

Prop AMM prices are not determined solely by the on-chain asset ratios in the pool but use a hybrid pricing model.

(1) Off-chain Calculation — Private Solver Decision Black Box When Jupiter aggregator initiates a quote request, the backend Solver is triggered, performing the following off-chain calculations within milliseconds: first, obtaining a benchmark price by subscribing to data streams from oracles like Pyth; the Solver runs its private algorithm model, which considers inventory risk, volatility, market sentiment, order flow toxicity prediction, and real-time costs; then, it outputs a deterministic quote including price, token amounts, and a valid timestamp. This entire process is completed on a server off-chain, with strategies, parameters, and calculations not publicly disclosed.

(2) On-chain Execution — Atomic, Protected Execution After off-chain calculation, the system seamlessly transitions to on-chain execution: the Solver submits the calculated quote via a signed transaction to the Solana network, which calls the Prop AMM’s on-chain smart contract to update its current valid quote. To ensure this latest quote is prioritized, the protocol may pay higher fees to validators like Jito—effectively bidding for block top execution rights—ensuring the quote is confirmed before regular user transactions, preventing arbitrage during quote updates. The aggregator compares all network quotes; if a Prop AMM wins, user trades are routed to its contract, which atomically verifies two key conditions: whether the current market price (usually still checked via oracle) remains within the quoted range, and whether the user’s trade size does not exceed the promised amount. Upon successful verification, assets paid by the user and the market maker’s quoted assets are exchanged within the same transaction and block, earning the spread for the market maker and delivering assets to the user. The entire process is either fully successful or fully fails, with no intermediate state.

2.3.2 Traffic Entry: Parasite on Aggregator (Jupiter)

Prop AMM adopts a “headless” model, with the only user entry point being leading aggregators like Jupiter. This resembles a parasitic relationship: Jupiter seeks the best prices for user trades and queries all liquidity sources (including Raydium, Orca, Meteora, and Prop AMMs).

All the bidding occurs within Jupiter’s routing algorithm, which is a millisecond-level price competition. In each quote request, the Prop AMM’s “brain” must respond within milliseconds. Only if its quote is at least equal to or better than public pools like Raydium/Orca/Meteora will the trade be routed to it. This design forces Prop AMMs to perform lean market making; any strategic mistake or technical delay can lead to traffic loss. Therefore, Prop AMMs and aggregators form a deeply symbiotic relationship: the former relies on the latter for user access, while the latter depends on the former for the best quotes to maintain market competitiveness.

2.3.3 MEV Resistance and IL Elimination

Risk management in Prop AMM is two-way: externally, it can actively defend against harmful MEV trading flows; internally, its proprietary capital model fundamentally transforms market making risk, converting uncontrollable impermanent loss (IL) into manageable inventory risk.

In public AMMs, liquidity is a static public facility, easily preyed upon by MEV bots. Prop AMM turns liquidity into an active “predator,” enabling a switch between offense and defense. First, by paying higher priority fees, its quotes are always prioritized, allowing price updates to be completed before attack bots can act, making “sandwich” attacks technically difficult. Second, the off-chain Solver can analyze transaction features in real-time. For “toxic orders” from known arbitrage addresses or exhibiting typical arbitrage patterns (e.g., following large trades, precise quantity calculations), the Solver can choose not to quote or provide a wide, unprofitable spread, effectively refusing service. Finally, Prop AMM often adopts JIT (Just-In-Time) liquidity strategies: after winning a trade on the aggregator, it immediately pulls assets from its treasury within the same atomic transaction or sequence of instructions to complete the swap, then withdraws. This minimizes exposure time, greatly reducing the window for MEV sandwich attacks.

Figure 2: HumidiFi Asset Composition

Additionally, Prop AMM has a significant advantage over traditional LPs in converting IL risk. In Prop AMM, the protocol uses its own capital, not aiming to maintain a fixed asset ratio, but to earn spreads through continuous buy and sell quotes. Asset inventory fluctuations (e.g., USDC increasing and SOL decreasing in a bull market) are a natural result of market making, not a loss. Therefore, the traditional concept of IL in AMMs does not apply here; instead, it is inventory risk. The protocol dynamically manages inventory via Solver algorithms, keeping it within target ranges, using methods such as active hedging and external balancing. When inventory imbalance becomes severe, the team can hedge on CEXs to quickly balance overall risk exposure. The final profit and loss of the protocol = total spread earned - market value fluctuations of inventory assets - operational costs. This is an active, professionally driven profit-loss model, not a passive IL borne by LPs unrelated to their decisions.

III. Market Landscape and Leading Players

The rise of Prop AMM is not a sporadic event but a profound transformation of the Solana DEX market structure. Since this year, the trading structure of Solana DEXs has shifted from meme-driven speculation to a stable SOL-Stablecoin base. Currently, SOL-Stablecoins contribute 60%-70% of Solana DEX trading volume, providing ample space for active market making strategies.

Figure 3: Solana DEX Trading Volume Structure

3.1 Prop AMM’s Niche in Solana Ecosystem

Prop AMM has grown rapidly this year, with several major Prop AMMs accumulating over $270 billion in trading volume. Currently, Prop AMMs account for 20%-40% of Solana’s weekly DEX trading volume.

Figure 4: Solana DEX: Prop AMM vs Traditional AMM (SOL-Stablecoin pairs, right)

Especially in SOL-Stablecoins, Prop AMMs have clearly occupied a strategic position at the top execution layer of Solana, with over 80% market share. In large trades (> $100k) of SOL-Stablecoins, Prop AMMs’ slippage advantage is nearly crushing traditional AMMs. This is because their liquidity depth is simulated based on CEXs rather than relying solely on on-chain TVL.

3.2 Main Players and Competition in Prop AMM

Within Prop AMM, due to differences in strategies and risk control models, the strength gap is very apparent.

Figure 5: Market Share of Solana Prop AMM Trading Volume

Since its emergence in late June, HumidiFi has become the largest DEX on Solana and Prop AMM within months, and the first to surpass $100 billion in cumulative trading volume. Currently, HumidiFi’s daily trading volume is around $1 billion, holding over 60% of the Prop AMM market share.

The core competitiveness of Prop AMM lies in the number of trading pairs it offers and the effectiveness of its proprietary and black-box strategies, with the tightness of spreads being key.

Figure 6: Number of Active Trading Pairs Served by Prop AMMs

Although Prop AMM’s market-making volume is mainly concentrated in SOL-Stablecoin core pairs, to compete for routing priority in aggregators and attract broader order flow, new protocols are trying to offer quotes on more long-tail assets. For example, Aquifer, launched in September, offers quotes for up to 190 trading pairs, surpassing the total offered by other Prop AMMs. ZeroFi, HumidiFi, and AlphaQ are also expanding their service boundaries, including top memecoins (like Fartcoin, USELESs) and cross-chain third-party assets (like MON, ZEC).

Of course, the core value of Prop AMM still lies in minimal spreads and stability.

Figure 7: Solana DEX Execution Price Difference (USDC-WSOL)

Reviewing the past month’s SOL/USDC trading data shows that Prop AMMs generally have lower average spreads but higher tail risks, mainly due to larger internal differences; while mainstream DEXs (like Meteora and PancakeSwap) have higher average spreads, they are more stable, reflecting stronger overall technical and operational capabilities.

Within Prop AMM, Tessera V and HumidiFi perform best in spread stability, with Tessera V compressing spreads to about 2 bps, and HumidiFi even below 1 bps—key reasons for their leading positions. GoonFi and ZeroFi are more volatile, with spreads exceeding 15 bps during fluctuations.

It is noteworthy that competition in the Prop AMM sector is intensifying. At least 8 firms are now engaged in dark pool operations, compared to only 2 at the beginning of the year. The token issuance logic of Prop AMMs has also become a valuation hotspot: HumidiFi, the largest dark pool, was the first to issue tokens, and doing so at the peak of business may yield higher market valuation and premiums. From a competitive perspective, later entrants issuing tokens first could undermine HumidiFi’s valuation, so leading firms are seizing the opportunity now to strengthen their market position. However, the economic model and utility of HumidiFi’s token (such as specific use cases and value accrual mechanisms) are still unclear, and its long-term value support and ecosystem positioning are uncertain, which is a focus of community and investor attention and discussion.

IV. Economic Model and Ecosystem Impact

The economic model of Prop AMM departs entirely from early DeFi token incentives or mining frameworks, instead following the profit logic of high-frequency trading (HFT) and professional market makers in traditional finance. Its impact on the Solana ecosystem will be a double-edged sword.

4.1 Profit Model

Prop AMM’s sustainability does not rely on external incentives but is based on its excellent execution efficiency and risk management capabilities.

Its main revenue comes from the spread—the tiny difference between buy and sell prices—similar to profits earned by CEX market makers. Relying on extreme execution and active hedging, even with narrow spreads, large daily trading volumes and very low operational costs could generate good profits. However, due to the closed nature of Prop AMM and lack of public data, some community members doubt whether HumidiFi, which compresses spreads to below 1 bps, can generate significant profits.

Additionally, to maintain optimal quoting position in aggregators, protocols may consider offering rebates via tokens or fee structures with specific order flow sources, which could be a major consideration for Prop AMM token issuance. Fundamentally, the final profit and loss depend on = total spread earned from market making - market value fluctuations of inventory assets - operational costs. This is a professionally driven active profit-loss model, not a passive IL borne by LPs unrelated to their decisions.

4.2 Double-Edged Sword for Solana Ecosystem

The impact of Prop AMM is complex; it is a classic trade-off between efficiency and fairness, openness and closedness, frequently seen in DeFi history.

For users, Prop AMM provides an almost zero-slippage, deep trading environment, especially for large traders, approaching CEX experience. It also concentrates market making from many retail participants into a few high-efficiency professional capital providers, theoretically improving overall market capital utilization and pricing accuracy.

However, behind this efficiency revolution and improved experience lie systemic challenges. Prop AMM captures the largest trading volume and high-profit spreads on main trading pairs, and as the service expands predictably, it will significantly reduce the revenue of ordinary LPs providing liquidity to public pools like Raydium and Orca. Long-term, if traffic and profits are continuously diverted, public liquidity pools may shrink and degenerate into only serving long-tail assets or extreme market conditions, potentially weakening the foundational openness and permissionless nature of DeFi.

The mystery of Prop AMM lies not only in the absence of a front-end but also in the anonymity of its team. If future liquidity and pricing power are concentrated in anonymous or institutional entities, new centralization nodes could form. Strategy errors, technical failures, or malicious acts could trigger severe on-chain market volatility. Moreover, DeFi’s innovation depends on protocol composability; Prop AMM’s closed interfaces and dynamic logic may hinder safe and reliable integration with other DeFi protocols (like lending, derivatives, automated vaults).

V. Conclusion and Outlook

The rise of Prop AMM is not just a technological optimization but a sign of the institutionalization and professionalization of on-chain trading on Solana, representing liquidity providers’ pursuit of ultimate capital efficiency.

In the future, Prop AMM may evolve into a professional market maker playing the role of an on-chain CEX for Solana. Its emergence is pushing Solana’s DEX market structure increasingly toward a traditional exchange model—order flow highly dependent on a few professional market makers. Solana is gradually becoming a transparent settlement layer akin to a CEX, with Prop AMM serving as the trading desks providing the lowest latency and best execution quality.

While users enjoy efficiency benefits brought by Prop AMM, this also compels traditional public AMMs to innovate, enabling ordinary LPs to access more automated market making strategies, achieving coexistence of public and institutional liquidity rather than marginalization. Traders may shift from relying solely on DEXs to using aggregators more frequently to maximize the benefits of Prop AMM’s minimal slippage and optimal execution.

VI. References

• Dune, https://dune.com/the_defi_report/prop-amms
• Blockworks, https://blockworks.com/analytics/humidifi/humidifi-tvl
• Dune, https://dune.com/sliceanalytics/solana-dex-activity
• Dune, https://dune.com/queries/6266421/9988032
• The Block, https://www.youtube.com/watch?v=LPQGl6Ju16U

(Click below for the full report)
[Gate Research Institute](https://www.gate.com/learn/category/research) is a comprehensive blockchain and cryptocurrency research platform providing in-depth content, including technical analysis, hot topics, market reviews, industry research, trend forecasts, and macroeconomic policy analysis.

Disclaimer Investing in cryptocurrency markets involves high risk. Users are advised to conduct independent research and fully understand the nature of assets and products before making any investment decisions. Gate is not responsible for any losses or damages resulting from such investment decisions.