• Prop AMM Innovation: Shift to dynamic, professional quoting via private vaults; market makers use proprietary off-chain algorithms to refresh parameters, enabling delivering very tight spreads (often under 5 bps) and minimal slippage even on large trades.

• Vs Traditional AMMs: Classic designs use passive fixed curves (x*y=k or concentrated liquidity), leading to high slippage, toxic flow, and MEV vulnerability due to predictable pricing.

• Dominance on Solana: Prop AMMs captured >50% of DEX volume on mature pairs (e.g., SOL/stables), often outperforming CEXs on spreads/slippage for trades under $1M.

• Solana-Exclusive Economics: High-frequency updates cost only a few thousand $/day on Solana (143–800 CU, <$0.002 each); on Ethereum, equivalent frequency would cost $60k–$300k/day, making it unviable.

• EVM Alternatives: Private liquidity exists via off-chain PMMs/RFQ (great for large clips but sacrifices on-chain composability, unlike Prop AMMs).

• Leading Players: HumidiFi (market leader), SolFi (Ellipsis Labs), TesseraV (Wintermute-backed); targeting institutional-scale $10M–$100M trades.

• Execution Metrics: Sub-1 bp spreads on SOL up to ~$500k, ~5 bps at $1M; average trade sizes 3–4x larger than public venues; sub-$100k clips now match TradFi-grade fills.

• Outlook: Pragmatic blend of TradFi market-making with on-chain settlement/composability, driving institutional adoption through superior execution but introduces centralization and counterparty risks over pure decentralization.

On-chain finance has long positioned itself as the vanguard of crypto's disruption of traditional finance. Beneath the surface, more profound innovations in on-chain market microstructure are positioning blockchain as a viable backend for global financial systems.

In the latter half of 2025, Solana's spot markets for mature assets quietly underwent a structural shift. Proprietary automated market makers (Prop AMMs) emerged as the dominant liquidity venues for liquid pairs like SOL/stables, capturing over 50% of DEX volume on these assets by year-end. This transition marks a departure from the passive, open-access designs that defined early DeFi toward a more professionalized, quote-driven model that prioritizes execution quality.

Prop AMMs have consolidated liquidity in high-volume corridors, delivering spreads and slippage profiles that rival or exceed those of centralized exchanges for sub-$1M clips. This development underscores Solana's unique position: its architectural advantages with low block times, cheap transactions, and predictable ordering, enable frequent on-chain parameter updates that make active market making economically viable in a fully on-chain settlement environment.

An AMM (Automated Market Maker) is a system that automatically prices and swaps tokens using a mathematical formula. An AMM uses the formula x * y = k to determine prices, where x and y are the amounts of the two assets in the pool, and k is a fixed constant.

Traditional AMMs, whether constant-product (x × y = k) or concentrated liquidity variants, rely on passive bonding curves to determine prices. In a constant-product pool, any trade shifts the reserves, introducing price impact. Concentrated liquidity improves capital efficiency by focusing provision within price ranges like Uniswap v3 with ticks mechanism but idle capital outside the active range and exposure to toxic flow remain inherent challenges.

These designs excel in permissionless, retail-driven environments but struggle with adverse selection. Public pricing is deterministic and temporally exploitable, inviting arbitrage and MEV attacks that erode LP returns and widen effective spreads for end users.

Prop AMMs address these shortcomings by replacing fixed curves with dynamic, model-based quoting. Liquidity resides in private vaults controlled by professional operators, who update external parameters (e.g., multipliers or concentration factors) on-chain multiple times per second. The bonding curve persists for deterministic settlement, but pricing reflects real-time inventory management, external oracles, and risk limits, mirroring TradFi market making without custodial intermediaries.

Access restriction to trusted aggregators insulates the venue from direct searcher inspection, neutralizing much of the MEV and toxic arbitrage that plagues public pools.

Let’s walk through the mechanics step by step.

The true power of Prop AMMs stems from their hybrid on/off-chain architecture, which enables active, professional-grade market making while preserving on-chain settlement and composability. Exact pricing formulas remain proprietary and typically closed-source (with bytecode often obfuscated to protect competitive edges), but the core process is well-understood from on-chain data and public implementations.

At a high level, Prop AMMs retain a bonding curve with a deterministic invariant for final settlement, but unlike classic AMMs, this invariant is dynamic: it depends on parameters continuously refreshed by the operator. The result is an actively managed curve that behaves like a TradFi market maker quoting in real time.

Step-by-Step Breakdown

1. Off-Chain Monitoring & Parameter Computation The professional market maker runs a sophisticated off-chain engine that continuously tracks external fair prices (CEX feeds, oracles like Pyth or Switchboard), current inventory and net delta exposure, volatility (realized and implied), risk thresholds (VaR-like limits, maximum position size), recent order flow patterns (to distinguish informed vs. retail flow), and available Just-In-Time (JIT) liquidity sources. Using these inputs, the engine computes optimal quoting parameters—such as curve multipliers, concentration factor around the oracle price, spread width, and maximum trade size.

2. Frequent On-Chain Parameter Updates The computed parameters are pushed on-chain via lightweight transactions. High priority fees ensure rapid inclusion, allowing leaders like HumidiFi to refresh 10–74 times per second. This "parameter spam" keeps the on-chain bonding curve tightly anchored to real-world fair value and concentrates effective liquidity precisely where needed.

3. Quote Request via Aggregator When a user initiates a swap (almost always through Jupiter, which routes ~90% of Solana volume), the aggregator queries available venues, including Prop AMMs, for the best available route. The Prop AMM returns a quote based on its latest on-chain parameters.

4. Dynamic Execution Evaluation Upon receiving the proposed trade, the Prop AMM program evaluates it against current risk models. It may accept outright, adjust pricing slightly within the transaction, incorporate JIT liquidity (sourcing or rebalancing inventory atomically in the same bundle), or reject/widen if the trade appears toxic. This provides an additional layer of adverse-selection protection.

5. Deterministic Settlement If the trade is filled, execution follows the updated bonding curve with full on-chain finality, preserving composability (e.g., bundling with flash loans or leverage in the same transaction).

Quotes indirectly rely on public oracles and on-chain parameters, which are visible. However, searchers cannot front-run or sandwich individual quotes profitably. Stale parameters are refreshed so frequently that arbitrage windows are minimal, and the off-chain engine can selectively reject toxic flow or widen spreads dynamically.

The true "secret sauce" varies by operator, but it primarily resides in the sophistication of their off-chain pricing engines rather than the on-chain mechanics alone.

All Prop AMMs retain a bonding curve with a deterministic invariant to ensure reliable on-chain settlement. However, unlike traditional AMMs where the invariant k remains fixed, here k (or its equivalent) is dynamic and fully parameterized. The exact formula is typically closed-source, rendering it effectively unknowable. The real competitive edge lies in the proprietary off-chain algorithms that compute optimal parameters based on oracles, inventory levels, and volatility. These parameters may include:

• External fair price references (CEX/DEX mid-prices, oracles such as Pyth or Switchboard)

• Current inventory and net delta exposure

• Realized and implied volatility

• Maximum exposure limits (VaR-like risk constraints)

• Order flow patterns (to distinguish informed from retail flow)

• Availability of Just-In-Time liquidity sources for real-time rebalancing

Ultimately, this architecture enables truly active pricing, resulting in ultra-tight spreads (often sub-5 bps), largely size-invariant slippage, and robust protection against adverse selection.

For a deeper dive into how Prop AMMs work and to better understand their underlying mechanisms, here is a great technical talk delivered by Chris Chang from Ghost at Breakpoint:

Leading implementations include HumidiFi, SolFi built by Ellipsis Labs, and TesseraV (Wintermute-backed). HumidiFi's co-founder has articulated ambitions to support institutional-scale clips ($10M–$100M) with spreads narrowed from ~30 bps historically to ~5 bps today.

Phoenix announcement from Ellipsis Labs highlighting extensions toward order-book hybrids:

Solana

@solana

BREAKING: @ellipsis_labs launches Phoenix Perpetuals, a new high performance Solana-native perps DEX

299

9:05 AM • Dec 11, 2025

Prop AMMs are a Solana-native phenomenon because frequent oracle updates (essential for tight quoting) are prohibitively expensive elsewhere. Even on Solana, these updates cost several thousand dollars daily; on Ethereum, they would be infeasible.

On Solana, leading operators like HumidiFi achieve up to 74 updates per second at 143–800 CU each, costing <$0.002 per refresh and several thousand dollars daily overall, viable given routed volume. On Ethereum, with average gas prices of 0.03–0.07 Gwei and typical simple transaction fees around $0.30–0.50 (21k gas for a transfer), a Prop AMM-like parameter update would consume roughly 30,000–50,000 gas, costing $0.01–0.05 per update at low congestion levels. At Solana-level frequencies (e.g., 74 updates/second, or ~6 million/day), daily costs would reach $60,000 to $300,000 (or higher during congestion). This is 10-100× more expensive than on Solana (a few thousand dollars/day at most), making the model entirely uneconomical and pushing ecosystems toward off-chain PMMs/RFQ instead. Even on L2s (Base, Optimism), costs remain significantly higher for such high-frequency on-chain activity.

Solana's combination of low transaction costs, high throughput, and aggregator concentration (Jupiter's dominant routing) creates a fertile environment: operators can maintain sharp prices without prohibitive overhead, while instant distribution to clean retail flow is guaranteed via a single integration.

While Prop AMMs have flourished uniquely on Solana, Ethereum and other EVM chains have developed analogous private liquidity models with distinct architectural trade-offs. The core economic driver remains the same: professional market makers seek to protect capital from toxic flow while delivering tight quotes. However, chain constraints dictate divergent implementations.

On Ethereum, private liquidity predominantly manifests through Private Market Makers (PMMs) operating largely off-chain. These entities provide quotes via RFQ systems or intent-based solvers, with presigned orders settled atomically on-chain. Over 90% of PMM order flow arrives via aggregators and intent integrations, insulating providers from direct searcher exposure.

The crucial advantage of Solana's Prop AMMs lies in full on-chain price determination. Quotes are computed directly within the program using dynamically updated parameters, making them instantly accessible to other smart contracts. This preserves composability: a trader can query a Prop AMM quote and bundle it with complex operations (e.g., flash loans, leverage positions) in a single atomic transaction.

RFQ/PMM systems, by contrast, route off-chain and break this atomicity. While exceptional for large, non-time-sensitive institutional clips, offering slippage-free execution with presigned guarantees, they cannot support composable DeFi strategies. A smart contract cannot "preview" an off-chain quote and conditionally execute downstream logic without introducing trust assumptions or multi-step workflows.

Efforts to port Prop AMM-like designs to EVM chains face prohibitive costs for frequent on-chain updates. While some proposals attempt to mitigate front-running risks, none match the native economic viability of Solana.

The measurable outcome of this microstructure evolution is dramatically improved execution quality. In TradFi, metrics like effective spreads and adverse selection (analogous to SEC Rule 605 disclosures) define venue competitiveness. Wholesalers routinely quote sub-1 bp on liquid equities. On Solana pre-2025, classic AMMs imposed structural floors: 5–9 bps on SOL pairs for small clips, widening nonlinearly due to fee tiers and depth constraints. Prop AMMs exhibit true market-maker behavior, spreads largely size-invariant, driven by inventory risk rather than passive curves.

Volume milestones underscore this shift: HumidiFi alone processed periods exceeding Binance's SOL/USD spot volume in Q4 2025, with collective Prop venues driving Solana DEXs to occasionally surpass major CEX aggregates. On MEV-related, protection arises from breaking predictability: restricted access and dynamic quoting render prices non-deterministic and non-exploitable in real time.

The rise of Proprietary AMMs stems from clear inefficiencies in traditional AMMs. These foundational designs embody core DeFi values: trustless automation, open composability, and true permissionless decentralization, which ensure security and accessibility. They are a great example of how open markets can run on code. Yet, due to poor LP profitability, driven by impermanent loss, toxic flow, and aggressive arbitrage, these models have been challenged by alternatives like CLOBs and Prop AMMs. The latter deliver superior efficiency with tighter spreads, deeper liquidity, and pricing that allows on-chain finance to compete with TradFi, but at the cost of new uncertainties and risks.

Key risks of Prop AMMs:

• Liquidity centralization: Dominated by a few professional (often pseudonymous) teams rather than open LP communities.

• Lack of transparency: No public frontend, private pricing logic, and heavy reliance on aggregators.

• Operational risks: Single market-maker dependency creates points of failure and potential adverse selection.

• Winner-takes-all dynamics: One dominant player can capture most volume, reducing venue diversity.

Today the ecosystem is evolving further with innovations aimed at enhancing classic AMMs while preserving greater transparency and decentralization, through protocol upgrades, hybrid models, and stronger MEV protections. These efforts seek a more sustainable balance between top-tier performance and DeFi’s founding principles.

Prop AMMs represent a pragmatic synthesis, importing TradFi's professionalized flow management while retaining on-chain settlement and (partial) composability. DeFi increasingly mirrors traditional market layers: retail aggregation, wholesale insulation, and sophisticated routing. Yet challenges persist. Liquidity efficiency has advanced dramatically, but we still need to ensure the ability to bring TradFi and institutional trading size on-chain and to be more resilient at scale. Centralization risks loom: a concentrated set of operators controls dominant venues, erecting high barriers via expertise and infrastructure. Off-chain pause capabilities introduce subtle counterparty considerations.

As we enter 2026, the critical path lies in the architectures of these models, which either democratize or consolidate an oligopoly. Solana’s microstructure shows that it is execution quality, not purist decentralization, that will ultimately attract sustainable institutional capital. But at what cost? Decentralization and composability remain central to the resilience and security of our on-chain infrastructure. The hybrid model emerging today may well represent the enduring equilibrium for on-chain spot markets. On-chain capital markets are looking TradFi in the eyes, and innovations have never been more prominent.

1. Chorus One – Solana Execution Quality Analysis (December 2025) Detailed thread comparing spreads, slippage, and behavior of Prop AMMs versus classic AMMs, with empirical data on SOL, BTC, and other pairs. https://x.com/ChorusOne/status/2003125074476294372

2. BarterSwap – Private Liquidity Blog Post In-depth analysis of private liquidity models, comparing Prop AMMs (Solana) with PMMs/RFQ systems (Ethereum), including diagrams and economic rationale. https://barterswap.xyz/blog/private-liquidity

3. 0xOptimus – Prop AMM Mechanics Thread (Core Explanation) Technical breakdown of Prop AMMs: dynamic invariants, oracle updates, and code examples (reference to Obric on Sui). https://x.com/0xOptimus/status/1981424092818399598

4. 0xOptimus – Composability vs. RFQ Comparison Thread Key explanation of why Prop AMMs preserve on-chain composability, unlike off-chain RFQ/PMM systems. https://x.com/0xOptimus/status/1985735558711185418

5. Kyle Samani (Multicoin Capital) – Commentary on Ellipsis Labs & Derivatives Progress VC perspective on the evolution of Prop AMM designs toward perpetuals and derivatives. https://x.com/KyleSamani/status/2000162347068067934

6. SolanaFloor – Quote from HumidiFi Co-Founder Kevin Pang Institutional ambitions: spread reduction from ~30 bps to ~5 bps, targeting $10M–$100M trade sizes.

• Prop AMM Innovation: Shift to dynamic, professional quoting via private vaults; market makers use proprietary off-chain algorithms to refresh parameters, enabling delivering very tight spreads (often under 5 bps) and minimal slippage even on large trades.

• Vs Traditional AMMs: Classic designs use passive fixed curves (x*y=k or concentrated liquidity), leading to high slippage, toxic flow, and MEV vulnerability due to predictable pricing.

• Dominance on Solana: Prop AMMs captured >50% of DEX volume on mature pairs (e.g., SOL/stables), often outperforming CEXs on spreads/slippage for trades under $1M.

• Solana-Exclusive Economics: High-frequency updates cost only a few thousand $/day on Solana (143–800 CU, <$0.002 each); on Ethereum, equivalent frequency would cost $60k–$300k/day, making it unviable.

• EVM Alternatives: Private liquidity exists via off-chain PMMs/RFQ (great for large clips but sacrifices on-chain composability, unlike Prop AMMs).

• Leading Players: HumidiFi (market leader), SolFi (Ellipsis Labs), TesseraV (Wintermute-backed); targeting institutional-scale $10M–$100M trades.

• Execution Metrics: Sub-1 bp spreads on SOL up to ~$500k, ~5 bps at $1M; average trade sizes 3–4x larger than public venues; sub-$100k clips now match TradFi-grade fills.

• Outlook: Pragmatic blend of TradFi market-making with on-chain settlement/composability, driving institutional adoption through superior execution but introduces centralization and counterparty risks over pure decentralization.

On-chain finance has long positioned itself as the vanguard of crypto's disruption of traditional finance. Beneath the surface, more profound innovations in on-chain market microstructure are positioning blockchain as a viable backend for global financial systems.

In the latter half of 2025, Solana's spot markets for mature assets quietly underwent a structural shift. Proprietary automated market makers (Prop AMMs) emerged as the dominant liquidity venues for liquid pairs like SOL/stables, capturing over 50% of DEX volume on these assets by year-end. This transition marks a departure from the passive, open-access designs that defined early DeFi toward a more professionalized, quote-driven model that prioritizes execution quality.

Prop AMMs have consolidated liquidity in high-volume corridors, delivering spreads and slippage profiles that rival or exceed those of centralized exchanges for sub-$1M clips. This development underscores Solana's unique position: its architectural advantages with low block times, cheap transactions, and predictable ordering, enable frequent on-chain parameter updates that make active market making economically viable in a fully on-chain settlement environment.

An AMM (Automated Market Maker) is a system that automatically prices and swaps tokens using a mathematical formula. An AMM uses the formula x * y = k to determine prices, where x and y are the amounts of the two assets in the pool, and k is a fixed constant.

Traditional AMMs, whether constant-product (x × y = k) or concentrated liquidity variants, rely on passive bonding curves to determine prices. In a constant-product pool, any trade shifts the reserves, introducing price impact. Concentrated liquidity improves capital efficiency by focusing provision within price ranges like Uniswap v3 with ticks mechanism but idle capital outside the active range and exposure to toxic flow remain inherent challenges.

These designs excel in permissionless, retail-driven environments but struggle with adverse selection. Public pricing is deterministic and temporally exploitable, inviting arbitrage and MEV attacks that erode LP returns and widen effective spreads for end users.

Prop AMMs address these shortcomings by replacing fixed curves with dynamic, model-based quoting. Liquidity resides in private vaults controlled by professional operators, who update external parameters (e.g., multipliers or concentration factors) on-chain multiple times per second. The bonding curve persists for deterministic settlement, but pricing reflects real-time inventory management, external oracles, and risk limits, mirroring TradFi market making without custodial intermediaries.

Access restriction to trusted aggregators insulates the venue from direct searcher inspection, neutralizing much of the MEV and toxic arbitrage that plagues public pools.

Let’s walk through the mechanics step by step.

The true power of Prop AMMs stems from their hybrid on/off-chain architecture, which enables active, professional-grade market making while preserving on-chain settlement and composability. Exact pricing formulas remain proprietary and typically closed-source (with bytecode often obfuscated to protect competitive edges), but the core process is well-understood from on-chain data and public implementations.

At a high level, Prop AMMs retain a bonding curve with a deterministic invariant for final settlement, but unlike classic AMMs, this invariant is dynamic: it depends on parameters continuously refreshed by the operator. The result is an actively managed curve that behaves like a TradFi market maker quoting in real time.

Step-by-Step Breakdown

1. Off-Chain Monitoring & Parameter Computation The professional market maker runs a sophisticated off-chain engine that continuously tracks external fair prices (CEX feeds, oracles like Pyth or Switchboard), current inventory and net delta exposure, volatility (realized and implied), risk thresholds (VaR-like limits, maximum position size), recent order flow patterns (to distinguish informed vs. retail flow), and available Just-In-Time (JIT) liquidity sources. Using these inputs, the engine computes optimal quoting parameters—such as curve multipliers, concentration factor around the oracle price, spread width, and maximum trade size.

2. Frequent On-Chain Parameter Updates The computed parameters are pushed on-chain via lightweight transactions. High priority fees ensure rapid inclusion, allowing leaders like HumidiFi to refresh 10–74 times per second. This "parameter spam" keeps the on-chain bonding curve tightly anchored to real-world fair value and concentrates effective liquidity precisely where needed.

3. Quote Request via Aggregator When a user initiates a swap (almost always through Jupiter, which routes ~90% of Solana volume), the aggregator queries available venues, including Prop AMMs, for the best available route. The Prop AMM returns a quote based on its latest on-chain parameters.

4. Dynamic Execution Evaluation Upon receiving the proposed trade, the Prop AMM program evaluates it against current risk models. It may accept outright, adjust pricing slightly within the transaction, incorporate JIT liquidity (sourcing or rebalancing inventory atomically in the same bundle), or reject/widen if the trade appears toxic. This provides an additional layer of adverse-selection protection.

5. Deterministic Settlement If the trade is filled, execution follows the updated bonding curve with full on-chain finality, preserving composability (e.g., bundling with flash loans or leverage in the same transaction).

Quotes indirectly rely on public oracles and on-chain parameters, which are visible. However, searchers cannot front-run or sandwich individual quotes profitably. Stale parameters are refreshed so frequently that arbitrage windows are minimal, and the off-chain engine can selectively reject toxic flow or widen spreads dynamically.

The true "secret sauce" varies by operator, but it primarily resides in the sophistication of their off-chain pricing engines rather than the on-chain mechanics alone.

All Prop AMMs retain a bonding curve with a deterministic invariant to ensure reliable on-chain settlement. However, unlike traditional AMMs where the invariant k remains fixed, here k (or its equivalent) is dynamic and fully parameterized. The exact formula is typically closed-source, rendering it effectively unknowable. The real competitive edge lies in the proprietary off-chain algorithms that compute optimal parameters based on oracles, inventory levels, and volatility. These parameters may include:

• External fair price references (CEX/DEX mid-prices, oracles such as Pyth or Switchboard)

• Current inventory and net delta exposure

• Realized and implied volatility

• Maximum exposure limits (VaR-like risk constraints)

• Order flow patterns (to distinguish informed from retail flow)

• Availability of Just-In-Time liquidity sources for real-time rebalancing

Ultimately, this architecture enables truly active pricing, resulting in ultra-tight spreads (often sub-5 bps), largely size-invariant slippage, and robust protection against adverse selection.

For a deeper dive into how Prop AMMs work and to better understand their underlying mechanisms, here is a great technical talk delivered by Chris Chang from Ghost at Breakpoint:

Leading implementations include HumidiFi, SolFi built by Ellipsis Labs, and TesseraV (Wintermute-backed). HumidiFi's co-founder has articulated ambitions to support institutional-scale clips ($10M–$100M) with spreads narrowed from ~30 bps historically to ~5 bps today.

Phoenix announcement from Ellipsis Labs highlighting extensions toward order-book hybrids:

Prop AMMs are a Solana-native phenomenon because frequent oracle updates (essential for tight quoting) are prohibitively expensive elsewhere. Even on Solana, these updates cost several thousand dollars daily; on Ethereum, they would be infeasible.

On Solana, leading operators like HumidiFi achieve up to 74 updates per second at 143–800 CU each, costing <$0.002 per refresh and several thousand dollars daily overall, viable given routed volume. On Ethereum, with average gas prices of 0.03–0.07 Gwei and typical simple transaction fees around $0.30–0.50 (21k gas for a transfer), a Prop AMM-like parameter update would consume roughly 30,000–50,000 gas, costing $0.01–0.05 per update at low congestion levels. At Solana-level frequencies (e.g., 74 updates/second, or ~6 million/day), daily costs would reach $60,000 to $300,000 (or higher during congestion). This is 10-100× more expensive than on Solana (a few thousand dollars/day at most), making the model entirely uneconomical and pushing ecosystems toward off-chain PMMs/RFQ instead. Even on L2s (Base, Optimism), costs remain significantly higher for such high-frequency on-chain activity.

Solana's combination of low transaction costs, high throughput, and aggregator concentration (Jupiter's dominant routing) creates a fertile environment: operators can maintain sharp prices without prohibitive overhead, while instant distribution to clean retail flow is guaranteed via a single integration.

While Prop AMMs have flourished uniquely on Solana, Ethereum and other EVM chains have developed analogous private liquidity models with distinct architectural trade-offs. The core economic driver remains the same: professional market makers seek to protect capital from toxic flow while delivering tight quotes. However, chain constraints dictate divergent implementations.

On Ethereum, private liquidity predominantly manifests through Private Market Makers (PMMs) operating largely off-chain. These entities provide quotes via RFQ systems or intent-based solvers, with presigned orders settled atomically on-chain. Over 90% of PMM order flow arrives via aggregators and intent integrations, insulating providers from direct searcher exposure.

The crucial advantage of Solana's Prop AMMs lies in full on-chain price determination. Quotes are computed directly within the program using dynamically updated parameters, making them instantly accessible to other smart contracts. This preserves composability: a trader can query a Prop AMM quote and bundle it with complex operations (e.g., flash loans, leverage positions) in a single atomic transaction.

RFQ/PMM systems, by contrast, route off-chain and break this atomicity. While exceptional for large, non-time-sensitive institutional clips, offering slippage-free execution with presigned guarantees, they cannot support composable DeFi strategies. A smart contract cannot "preview" an off-chain quote and conditionally execute downstream logic without introducing trust assumptions or multi-step workflows.

Efforts to port Prop AMM-like designs to EVM chains face prohibitive costs for frequent on-chain updates. While some proposals attempt to mitigate front-running risks, none match the native economic viability of Solana.

The measurable outcome of this microstructure evolution is dramatically improved execution quality. In TradFi, metrics like effective spreads and adverse selection (analogous to SEC Rule 605 disclosures) define venue competitiveness. Wholesalers routinely quote sub-1 bp on liquid equities. On Solana pre-2025, classic AMMs imposed structural floors: 5–9 bps on SOL pairs for small clips, widening nonlinearly due to fee tiers and depth constraints. Prop AMMs exhibit true market-maker behavior, spreads largely size-invariant, driven by inventory risk rather than passive curves.

Volume milestones underscore this shift: HumidiFi alone processed periods exceeding Binance's SOL/USD spot volume in Q4 2025, with collective Prop venues driving Solana DEXs to occasionally surpass major CEX aggregates. On MEV-related, protection arises from breaking predictability: restricted access and dynamic quoting render prices non-deterministic and non-exploitable in real time.

The rise of Proprietary AMMs stems from clear inefficiencies in traditional AMMs. These foundational designs embody core DeFi values: trustless automation, open composability, and true permissionless decentralization, which ensure security and accessibility. They are a great example of how open markets can run on code. Yet, due to poor LP profitability, driven by impermanent loss, toxic flow, and aggressive arbitrage, these models have been challenged by alternatives like CLOBs and Prop AMMs. The latter deliver superior efficiency with tighter spreads, deeper liquidity, and pricing that allows on-chain finance to compete with TradFi, but at the cost of new uncertainties and risks.

Key risks of Prop AMMs:

• Liquidity centralization: Dominated by a few professional (often pseudonymous) teams rather than open LP communities.

• Lack of transparency: No public frontend, private pricing logic, and heavy reliance on aggregators.

• Operational risks: Single market-maker dependency creates points of failure and potential adverse selection.

• Winner-takes-all dynamics: One dominant player can capture most volume, reducing venue diversity.

Today the ecosystem is evolving further with innovations aimed at enhancing classic AMMs while preserving greater transparency and decentralization, through protocol upgrades, hybrid models, and stronger MEV protections. These efforts seek a more sustainable balance between top-tier performance and DeFi’s founding principles.

Prop AMMs represent a pragmatic synthesis, importing TradFi's professionalized flow management while retaining on-chain settlement and (partial) composability. DeFi increasingly mirrors traditional market layers: retail aggregation, wholesale insulation, and sophisticated routing. Yet challenges persist. Liquidity efficiency has advanced dramatically, but we still need to ensure the ability to bring TradFi and institutional trading size on-chain and to be more resilient at scale. Centralization risks loom: a concentrated set of operators controls dominant venues, erecting high barriers via expertise and infrastructure. Off-chain pause capabilities introduce subtle counterparty considerations.

As we enter 2026, the critical path lies in the architectures of these models, which either democratize or consolidate an oligopoly. Solana’s microstructure shows that it is execution quality, not purist decentralization, that will ultimately attract sustainable institutional capital. But at what cost? Decentralization and composability remain central to the resilience and security of our on-chain infrastructure. The hybrid model emerging today may well represent the enduring equilibrium for on-chain spot markets. On-chain capital markets are looking TradFi in the eyes, and innovations have never been more prominent.

1. Chorus One – Solana Execution Quality Analysis (December 2025) Detailed thread comparing spreads, slippage, and behavior of Prop AMMs versus classic AMMs, with empirical data on SOL, BTC, and other pairs. https://x.com/ChorusOne/status/2003125074476294372

2. BarterSwap – Private Liquidity Blog Post In-depth analysis of private liquidity models, comparing Prop AMMs (Solana) with PMMs/RFQ systems (Ethereum), including diagrams and economic rationale. https://barterswap.xyz/blog/private-liquidity

3. 0xOptimus – Prop AMM Mechanics Thread (Core Explanation) Technical breakdown of Prop AMMs: dynamic invariants, oracle updates, and code examples (reference to Obric on Sui). https://x.com/0xOptimus/status/1981424092818399598

4. 0xOptimus – Composability vs. RFQ Comparison Thread Key explanation of why Prop AMMs preserve on-chain composability, unlike off-chain RFQ/PMM systems. https://x.com/0xOptimus/status/1985735558711185418

5. Kyle Samani (Multicoin Capital) – Commentary on Ellipsis Labs & Derivatives Progress VC perspective on the evolution of Prop AMM designs toward perpetuals and derivatives. https://x.com/KyleSamani/status/2000162347068067934

6. SolanaFloor – Quote from HumidiFi Co-Founder Kevin Pang Institutional ambitions: spread reduction from ~30 bps to ~5 bps, targeting $10M–$100M trade sizes. https://x.com/SolanaFloor/status/1999464575394283694

7. Solana Official Account – Ellipsis Labs Phoenix Announcement Repost Ecosystem context and highlight of hybrid order-book innovation from Ellipsis Labs. https://x.com/solana/status/1999051219768426991

8. Jupiter Exchange – DTF Launch Featuring HumidiFi Integration Illustration of aggregator dominance in routing flow to Prop AMMs. https://x.com/JupiterExchange/status/1983649251017290244

9. Coinbase Ventures – Ideas We Are Excited For in 2026 Institutional outlook on aggregator-insulated liquidity models and their role in broader on-chain adoption. https://www.coinbase.com/fr-fr/blog/Coinbase-Ventures-Ideas-we-are-excited-for-in-2026

10. Dune Analytics – Prop AMMs Volume Dashboard (The Defi Report) Volume and market share data for Prop AMMs on Solana throughout 2025. https://dune.com/the_defi_report/prop-amms

11. Uniswap Documentation – How Uniswap Works (v2 Overview) Canonical reference for the constant-product formula x × y = k and traditional AMM mechanics. https://docs.uniswap.org/contracts/v2/concepts/protocol-overview/how-uniswap-works

12. Flashbots GitHub – Global Storage Smart Contract Proposal Technical proposal for enabling Prop AMM-like mechanisms on Ethereum via priority lanes. https://github.com/flashbots/global-storage-smart-contract

13. Helius - Solana’s Proprietary AMM Revolution https://www.helius.dev/blog/solanas-proprietary-amm-revolution?referrer=grok.com