Core Architecture and Positioning of DeFi's Top Money Markets
Castle Lending Series: Part I
Introduction
Onchain lending is having a new growth wave. With around $80 billion in assets now locked across money markets, it stands as the largest vertical in DeFi. But this growth is more than just a return of capital onchain. It signals a shift in how lending is being built, with new architectures, more innovative risk models, and radically different user experiences emerging across the stack.
After a sluggish 2023-24, lending protocols have once again become a driving force in DeFi.
Capital efficiency is now the competitive edge. Protocols are racing to optimise LTVs, enable looping, and deploy productive collateral.
Modular, vault-based designs are replacing legacy pool systems, offering greater flexibility and faster iteration.
Institutions are entering the space, from tokenised treasuries on Aave Horizon to BTC-backed loans via Coinbase and Morpho.
DEX-Lending convergence is taking shape, with protocols like Fluid and Euler blending borrowing and trading into a single liquidity layer.
This evolution is not just technical; it is strategic. The lending sector is becoming more competitive, with various protocols taking radically different paths toward scalability, composability, and risk isolation.
A Framework for Comparison
This report marks the beginning of a multi-part series on the technical aspects of onchain lending. We begin by examining four protocols that best represent this new era: Aave, Morpho, Euler, and Fluid.
To structure our analysis, we focus on the three dimensions used throughout the report:
Market Positioning: Who is the protocol built for? Where is it deployed? How much value does it command, and what ecosystem relationships shape its strategy?
Core Features and Architectural Design: What is the protocol’s foundational structure? Is it modular, pooled, or vault-based? What features define the user experience and capital efficiency?
Future Roadmap: What upgrades, expansions, or paradigm shifts are on the horizon? How do they build on the existing architecture, and what direction is the protocol heading?
This structure helps draw clear lines between different lending models, highlighting how each project balances tradeoffs between risk, flexibility, and growth.
Comparative Overview
Aave
Aave is DeFi’s default money market, defined by depth and trust. It commands over half of all lending value locked, making it the venue of choice for both institutions and retail users. Its ability to scale liquidity across chains while maintaining conservative risk management has created a durable moat, reinforced by continuous innovation. From its dual-market structure in v3 to the upcoming Liquidity Hub and Spokes in v4, Aave is not just defending its position; it is evolving into the foundational infrastructure layer for onchain capital markets.
Market Positioning
Aave stands as the largest lending protocol by Total Value Locked (TVL), significantly outpacing its nearest competitors. At the time of writing, Aave controls ~$40 billion in TVL, which is more than five times that of Morpho, the second-largest protocol in the category. Its scale provides a strong competitive moat, enabling it to attract institutional capital while remaining highly accessible to retail users seeking passive yield opportunities.
Aave’s growth has been consistent over time, with its TVL recently hitting an all-time high. This growth can be attributed to a combination of early-mover advantage, continuous product development, and its broad multi-chain strategy.
The protocol is currently deployed across 19 chains, though the vast majority of its liquidity (over 85%) remains concentrated on Ethereum. Among Layer 2 (L2) networks, Arbitrum leads, followed by Linea and Base. This positioning reflects a strategy that prioritises Ethereum’s depth and security while selectively extending to scaling solutions with growing user activity.
Aave’s liquidity depth makes it especially attractive to High Net Worth Individuals (HNWIs) and institutions. These users can borrow large amounts without triggering significant spikes in borrowing rates. This dynamic was validated by the Ethereum Foundation’s deployment of 30,800 ETH into Aave in February 2025.
Core Features and Architectural Design
At its core, Aave operates on a pooled liquidity model, where users deposit assets into shared liquidity pools. Depositors receive aTokens representing their position, which accrue interest over time and can be used as collateral to borrow against.
Aave v3 introduced two distinct lending markets, each serving a different strategic function:
Aave Prime is optimised for capital efficiency and supports only blue-chip assets. Its defining feature is Efficiency Mode, which allows highly correlated assets like wstETH and WETH to be used with an LTV of up to 95% and a liquidation threshold of 96.5%. This market currently has a size of around $2 billion.
Aave Core is broader, with a market size exceeding $54 billion, supporting a wide range of assets. It prioritises risk control by implementing Isolation Mode, which caps exposure to higher-risk assets and enforces stricter LTV limits, generally below 85%.
This dual-market structure reflects Aave’s effort to balance capital efficiency and systemic safety. However, it also creates capital fragmentation, as liquidity cannot flow freely between Prime and Core markets. Resolving this inefficiency is a central focus of the upcoming v4.
The following diagram illustrates the various components within Aave, which function together efficiently to make Aave a leading money market.
Liquidity Pools and Reserves
Each token market on Aave is configured as a reserve, governed by DAO-approved parameters:
Loan-to-Value (LTV): the percentage of collateral value that can be borrowed. For example, an 85% LTV allows a $1,000 ETH deposit to back an $850 loan.
Liquidation Threshold: the collateral value at which liquidation becomes possible, typically set slightly above the LTV to create a safety buffer.
Borrowing Enabled: determines whether a reserve can be borrowed.
Caps: limits on how much can be supplied or borrowed, which prevent systemic overexposure to volatile assets.
Interest Rate Model (IRM): dynamically adjusts borrowing costs based on utilisation. High utilisation drives rates up, incentivising repayment or more supply.
Together, these parameters provide Aave with an adaptive and decentralised risk framework, featuring continuous governance-driven updates that reflect market conditions.
Aave Umbrella
Aave’s Umbrella system (previously the Safety Module) provides a critical safeguard. Users can stake aTokens or GHO to earn rewards, while also acting as a backstop against bad debt. If insolvency occurs, staked assets can be slashed to cover losses. The DAO has guaranteed up to $100,000 of initial bad debt coverage, creating a meaningful insurance layer for the protocol.
Liquidations are another cornerstone of Aave’s resilience. Every borrow position is monitored by a Health Factor (HF), defined as:
HF = (Collateral Value × Liquidation Threshold) / Borrowed Value
If the HF drops below 1, the position is eligible for liquidation. For example, depositing $1,000 in ETH and borrowing $800 USDC at an 88% liquidation threshold gives an HF of 1.1. If ETH falls to $900, the HF drops to 0.99, triggering liquidation.
Liquidators are incentivised by being able to repay up to 50% of a borrower’s debt and receiving collateral at a discount. This system keeps Aave solvent, while creating market opportunities for arbitrageurs and professional liquidation bots.
Efficiency and Isolation Modes
Efficiency Mode (E-Mode) maximises borrowing power for tightly correlated assets like staked ETH derivatives. This feature enables capital-efficient strategies such as recursive lending or leveraged staking.
Isolation Mode mitigates the risks associated with newer or volatile tokens by restricting what can be borrowed against them and limiting their exposure caps.
These mechanisms reflect Aave’s philosophy: expand cautiously, but innovate aggressively in capital efficiency where risk is quantifiable.
Future Roadmap
The upcoming Aave v4 (launching in Q4 this year) is designed to address capital fragmentation and further advance the protocol as a modular lending infrastructure.
The centrepiece is the Liquidity Hub, which aggregates liquidity across Prime and Core Markets. Instead of being locked in silos, capital will flow dynamically to where it is needed, improving efficiency and reducing idle deposits.
A complementary innovation is Spokes, specialised market instances that plug into the Liquidity Hub. Each Spoke can operate with its own parameters, such as supporting long-tail assets, using custom interest rate models, or experimenting with new governance setups. This design allows innovation at the edges without compromising the safety of the core system.
Aave is also moving into Real-World Asset (RWA) lending through its Horizon markets. By accepting tokenised collateral such as Money Market Funds (MMFs) from issuers like Superstate and Centrifuge, Aave is positioning itself as a bridge between traditional institutions and DeFi. This is part of a broader strategy to capture institutional flows, complementing its already strong base of onchain users.
Finally, enhancements to governance and liquidation tooling are expected to provide the DAO with more precise control over risk settings, enabling more automated and capital-efficient liquidations.
Taken together, these upgrades mark Aave’s evolution from a single dominant protocol into a platform layer for onchain capital markets. By unifying liquidity, introducing modularity, and opening doors for RWAs, Aave v4 is laying the groundwork to remain the cornerstone of DeFi lending in the years ahead.
Morpho
Morpho has redefined what it means to build a lending protocol by adopting a modular and permissionless approach. Once an optimiser layer on Aave and Compound through Morpho V0, Morpho has since evolved into Morpho V1, a lending network that enables permissionless, immutable market creation. Its unique architecture, combined with integrations like Coinbase’s cbBTC loans, has made Morpho the leading challenger to Aave. By prioritising immutability, flexibility, and composability, Morpho positions itself as the ecosystem’s most adaptable money market.
Market Positioning
Morpho is the second-largest lending protocol after Aave, with TVL climbing rapidly since the launch of Morpho V1 in early 2024. It currently holds around $7 billion in TVL, with a total market size (supplied assets) surpassing $11 billion. This represents remarkable growth compared to Morpho V0, which reached a peak of $2 billion and has since been deprecated.
Morpho’s most important strategic move has been its close alignment with Base. It is now the largest protocol by TVL on Base, with around 25% of the network’s liquidity. Building on Base has provided it with a strong foothold in the ecosystem and helped fuel rapid growth. Morpho is live across 29 chains, reflecting a “Morpho Everywhere” strategy that aims to deploy its contracts broadly for maximum composability.
The protocol is also powering Coinbase crypto-backed loans. Users can deposit BTC on Coinbase, where it is tokenised as cbBTC and moved onchain to Morpho, enabling them to borrow USDC. This integration has proven significant: Coinbase users have borrowed over $700 million USDC with $1.3B+ cbBTC collateralised, which accounts for more than 15% of all borrowed funds on Morpho.
Source: https://dune.com/queries/4901055/8114608
Beyond Coinbase, Morpho has attracted integrations from exchanges and Fintechs like Gemini, Crypto.com, Bitpanda, Ledger, Trust Wallet and more. Its open design allows other DeFi protocols, such as Compound, Moonwell Lending, and Spark DAO, to deploy vaults and markets directly on Morpho, either using its own UI or building custom frontends via providers like Paperclip. This ecosystem-led approach differentiates Morpho from Aave’s more centralised design and aligns it with DeFi’s permissionless ethos.
Core Features and Architectural Design
At its foundation, Morpho V1 is built as an immutable base layer that supports the creation of permissionless markets. Once created, the market is immutable, and no one can change it. The diagram below illustrates the various components that interact within Morpho. This architecture doesn’t include Vaults V2 and Markets V2, and only considers their V1. Morpho components include Vaults, Markets, Public Allocators and more.
Markets
Each Morpho market operates independently and has five parameters:
Collateral Asset: the asset posted as collateral
Loan Asset: the asset borrowed against the collateral
Liquidation Loan-to-Value (LLTV): the maximum % of collateral value that can be borrowed before liquidation
Oracle: provides price data for the assets
Interest Rate Model (IRM): determines how borrowing rates respond to utilisation
This structure keeps each market isolated, which limits systemic contagion risk.
Vaults
Morpho introduces Vaults built on top of markets, offering automated asset curation and risk optimisation. Each Vault has a single loan asset but can spread deposits across multiple underlying markets. This design allows users to earn optimised yields without having to manage individual lending positions.
Vaults also introduce role-based governance:
Owner: appoints curators and allocators
Curator: decides which markets the vault allocates funds to
Allocator: manages deposits, withdrawals, and rebalancing
Guardian: a safety role that can veto malicious actions
This multi-role framework enables Morpho Vaults to be flexibly managed, striking a balance between decentralisation and risk oversight.
Morpho Curators
Morpho Curators are crucial components of Morpho Vaults, acting as Asset Managers who optimise risk-adjusted returns for depositors by curating the Vault’s Asset Allocation Strategy and setting exposure limits or caps.
Their responsibilities include:
Strategic Curation: Defining the vault’s overall investment thesis, including selecting permissible protocols, markets, or asset types for allocation.
Risk Parameterisation: Setting and maintaining risk boundaries for the vault by defining exposure limits (caps) to various risk factors to protect depositor capital.
Addressing Borrower Needs: Ensuring competitive rates, reliable liquidity during market stress, adequate capital support for growing positions, and quick adaptations to changing market conditions.
Addressing Depositor Needs: Focusing on risk mitigation (zero tolerance for bad debt), competitive yields, robust security and continuous collateral liquidity oversight, and transparent, ongoing risk assessments.
There are multiple curators currently in Morpho, including Gauntlet, SteakHouse, MEV Capital, Smokehouse, and others.
Morpho Lite
To make deployment easier, Morpho developed Morpho Lite, a streamlined version of its front-end app that supports the core functions: Earn, Borrow, and Rewards. Morpho is also easily integrated into other DeFi frontends, allowing it to be quickly deployed across multiple chains, such as Hyperliquid, World, Plume, and others.
This has been a critical part of Morpho’s expansion strategy, allowing it to integrate seamlessly into different ecosystems and spread more rapidly.
Integrator Ecosystem
Morpho’s flexible architecture has attracted a wide range of integrations:
Fintechs and exchanges such as Coinbase, Gemini, Crypto.com, Ledger and Trust Wallet integrate Morpho Markets and Morpho Vaults for custom lending/borrowing use cases for their millions of end users.
Compound, Moonwell Lending, and Spark DAO are active integrators that deploy and manage Morpho Vaults.
Providers like Paperclip allow protocols to build custom frontends that plug directly into Morpho’s infrastructure.
On HyperEVM, Felix Protocol offers vanilla markets powered by Morpho’s contracts.
This demonstrates Morpho’s “Morpho Everywhere” strategy: by embedding itself into multiple consumer applications, protocols and chains, it maximises its reach and composability.
Liquidation Mechanisms
Morpho supports two forms of liquidation:
Standard Liquidation: positions exceeding the LLTV can be liquidated, with liquidators repaying debt in exchange for collateral plus a bonus
Pre-Liquidation (Auto-Deleverage): an opt-in mechanism where smaller liquidations occur once a position crosses a pre-LLTV threshold but is not yet fully at risk
Example: A user deposits $10,000 cbBTC and borrows $7,000 USDC in a market with an LLTV of 80% and a pre-LLTV of 75%. If the collateral value drops to $9,200, pushing LTV near 75%, pre-liquidation triggers partial repayment, reducing the risk of a hard liquidation later.
This system smooths liquidations and reduces volatility, which is especially valuable for borrowers using long-tail collateral.
Interest Rate Model
Morpho employs an AdaptiveCurveIRM designed to keep utilisation near a target of 90%. The model consists of two mechanisms:
Curve Mechanism: prevents extreme utilisation by adjusting rates sharply at the edges. If utilisation hits 100%, rates spike to around 16%. At 0%, rates fall near 1%.
Adaptive Mechanism: adjusts gradually to maintain balance. If utilisation lingers too low (say 45%), borrowing costs fall, encouraging demand. If utilisation is high (95%), rates increase, incentivising repayment.
This dynamic IRM makes Morpho suitable for permissionless market creation while protecting liquidity providers.
While the IRM dynamically adjusts borrowing costs based on utilisation, Morpho curators play a vital role in indirectly influencing these rates through their liquidity management strategies. By optimising the allocation of capital to different markets within a vault, curators can impact market utilisation, thereby affecting the rates determined by the AdaptiveCurveIRM. Responsible curation is required to ensure that vaults maintain competitive yields for depositors while providing efficient liquidity and acceptable fees for borrowers.
Public Allocators
One challenge with isolated markets is that they can fragment liquidity. Morpho addresses this issue with Public Allocators, which enable liquidity to be drawn from multiple markets into a single loan. This effectively gives borrowers access to deep, aggregated liquidity while maintaining risk isolation across markets.
Future Roadmap
Morpho’s next protocol is Morpho V2, which introduces fixed-rate fixed-duration lending and complements variable rates of Morpho V1. Morpho V2 is also intent-based. This represents a significant departure from traditional pool models. Instead of simply depositing into markets, lenders and borrowers will be able to post offers with specific terms, such as:
Fixed interest rates and durations
Multi-collateral positions
Portfolio-based collateral, including Real-World Assets (RWAs)
This design enables more efficient price discovery and expands lending to a broader range of asset classes.
Morpho V2 also introduces cross-chain lending, allowing users to provide liquidity on one chain while borrowers access it on another. This could dramatically increase capital efficiency and position Morpho as a multi-chain liquidity hub.
In addition, Vaults V2 will expand customisation and automation features, giving DAOs, treasuries, and protocols more tools to manage capital through Morpho.
The role of curators will also evolve in V2, with the following expanded responsibilities:
Appointing Managers: In Vault v2, Curators are responsible for appointing Allocators who actively manage the portfolio within the established risk framework.
Setting Core Vault Mechanics: In Vault v2, Curators also control crucial mechanisms, such as fees and compliance gating.
Taken together, these upgrades reinforce Morpho’s identity as the most modular and permissionless lending protocol in DeFi. By enabling both open market creation and intent-based lending, it positions itself not only as a competitor to Aave but also as a flexible platform that can underpin a wide array of lending strategies across chains.
Euler
Euler is the programmable credit layer of onchain finance. With its V2 relaunch, Euler moved beyond pooled lending and rebuilt itself as a vault-native system where builders, lenders, and borrowers each have a role. Its architecture empowers developers and DAOs to create vaults that collateralise almost any kind of digital asset, while still supporting direct borrowing and lending for users. This makes Euler less of a single money market and more of a modular platform for credit, one that balances composability with risk isolation.
Market Positioning
Euler V2 was launched in September 2024, following extensive audits, and positioned itself not only as a lending protocol but also as an infrastructure layer. The protocol has reached approximately $1.5 billion in TVL across 12 chains, with Ethereum and Avalanche hosting the majority of the liquidity. This spread reflects a deliberate strategy: keep Ethereum as the anchor for deep, secure markets, while using other EVM chains to expand reach.
Where Aave dominates through scale, and Morpho gained market share through key integrations and partnerships, Euler has carved out a niche around programmability. Its architecture is designed to empower builders, DAOs, and advanced users to create customised credit markets. Vaults can be launched permissionlessly through the Euler Vault Kit (EVK), each configured with its own oracle, interest rate model, and risk parameters. The Ethereum Vault Connector (EVC) links these vaults together, allowing collateral in one to support borrowing in another and enabling cross-vault strategies that would not be possible in a pooled system.
Euler currently operates two major market instances on Ethereum. Euler Prime focuses on conservative, blue-chip assets such as ETH and wstETH, providing a foundation for safe, predictable borrowing and lending. Euler Yield supports more experimental strategies, including markets for yield-bearing tokens such as Pendle PTs. This dual structure signals Euler’s ambition to appeal to both institutions that demand stability and to DeFi-native users seeking more complex, yield-optimised positions.
In positioning itself this way, Euler is not competing directly with Aave’s liquidity moat or Morpho’s optimisation engine. Instead, it appeals to those who value customisation, composability, and programmability as the foundation of their credit markets.
Core Features and Architectural Design
Euler V2 is built around ERC-4626 vaults as its core primitive. Each vault is isolated, holding a single underlying asset and enforcing its own borrowing caps, liquidation thresholds, oracles, and interest rate model. This design provides strong risk isolation, since failures in one vault cannot cascade through the system, while allowing vaults to be configured flexibly for different use cases.
The diagram below provides an overview of these components and their functions, which make Euler an efficient lending protocol.
The Euler Vault Kit (EVK)
The EVK provides the framework for creating and managing vaults. EVault contracts handle deposits, borrowing, and accounting, while oracles supply real-time pricing. Interest rate models can be selected and tailored to each vault. ProtocolConfig defines global fee structures, and DTokens represent borrower debt transparently onchain. This modular design allows DAOs or developers to launch vaults that reflect their own risk profile, from tightly capped ETH vaults to more experimental configurations for yield-bearing assets.
Source: Euler Explorer
Markets
From these vaults, Euler assembles its markets, which connect a collateral asset to a borrowable asset under defined rules. Borrowing capacity depends on each market’s liquidation threshold, oracle feed, and IRM. The isolation of markets ensures that problems in one do not threaten the system as a whole, while allowing different markets to adopt very different strategies. This is what makes Euler attractive to both conservative institutions and DeFi-native builders.
The Ethereum Vault Connector (EVC)
The Ethereum Vault Connector (EVC) links isolated vaults into a composable system. Collateral in one vault can back borrowing in another, making it possible to construct multi-leg strategies directly within Euler. The EVC supports batch transactions, letting users combine actions such as deposit, borrow, and swap into a single transaction. It also allows for programmable stop-loss conditions, enabling users to automate deleveraging or repayment when markets move against them.
Interest Rate Models and Dynamic Utilisation
Euler’s vaults use modular Interest Rate Models (IRMs) rather than a single global curve. Rates adjust dynamically based on each vault’s utilisation, aiming to optimise liquidity and balance supply with demand. Stablecoin vaults typically employ conservative curves, whereas long-tail or yield-bearing assets can be paired with smoother, adaptive models.
Liquidations and Risk Management
Euler uses the familiar Health Factor model:
HF = (Collateral Value × Liquidation Threshold %) / Borrowed Value
Positions with HF below 1 become liquidatable. Liquidators repay debt and receive collateral at a discounted rate, thereby preserving solvency. By default, Euler uses bad debt socialisation, meaning that if a vault accrues bad debt, the liability is spread across depositors. Vault governors, however, can disable this feature and choose alternative mechanisms. This adds another layer of flexibility, ensuring that vaults can be designed to handle risk in different ways. Euler also maintains an open-source liquidation bot, which Euler Labs operate.
EulerSwap
Euler has already begun rolling out EulerSwap, an integrated automated market maker that lets vault collateral double as trading liquidity. This reduces capital fragmentation by allowing assets to earn trading fees while remaining usable as collateral. Unlike traditional AMMs, EulerSwap deploys liquidity just in time for trades, improving efficiency. It is live in beta on select networks and accessible through major aggregators such as CoWSwap and ParaSwap, with deeper integration planned.
Hooks and Programmability
A defining feature of Euler is its support for hooks, external contracts that vault conditions can trigger. Hooks can enforce custom logic, automate actions, or even block transactions that do not meet requirements. For example, a hook could automatically hedge collateral, trigger deleveraging, or impose compliance restrictions. This programmability is central to Euler’s identity as a credit platform rather than a single money market.
User-Facing Tools
Although Euler’s architecture is optimised for builders, it also includes features that make advanced strategies accessible to individuals. Multiply automates looping by combining deposit, borrow, swap, and redeposit into a single transaction, giving users leveraged exposure without manual effort. Earn Vaults provide passive yield for users who prefer simplicity, with risk curators managing rebalancing, liquidity, and strategy allocation. Spy Mode adds transparency by allowing portfolios to be viewed from another user’s perspective, a helpful feature for DAOs monitoring treasuries. Reward Streams enable new markets to bootstrap adoption by distributing incentives to depositors and borrowers in a permissionless manner.
Future Roadmap
Scaling EulerSwap
EulerSwap is now live in beta, with vault liquidity seeded on select networks and integrations through major DEX aggregators, including CoWSwap and ParaSwap. The roadmap focuses on expanding its reach: supporting more assets, deepening liquidity, and rolling out just-in-time mechanics more broadly. While EulerSwap has already recorded significant trading activity in beta, its role is expected to grow as it becomes more tightly integrated with vaults and markets.
Advancing Hooks
Hooks are already part of Euler’s design, but the team has signalled plans to make them more powerful. Future upgrades could allow developers to implement custom risk engines, automated hedging modules, or compliance rules directly inside vaults. By extending hooks, Euler moves closer to its ambition of becoming a programmable platform where lending logic can be tailored to any use case.
Cross-Chain Expansion
Euler is live on multiple EVM chains, with Ethereum and Avalanche currently hosting most of its liquidity. Expansion to additional chains is on the agenda, but each deployment is expected to maintain isolated risk parameters while leveraging the same EVK and EVC framework. This modular approach ensures that growth across chains does not compromise system safety.
Institutional Adoption
Euler has not published a formal roadmap for institutional features, but community discussions suggest interest in permissioned vaults, compliance wrappers, and governance upgrades. These would make the system more appealing to DAOs and regulated entities that require stricter risk and reporting standards. The timeline for these developments remains open, but the intent aligns with broader DeFi trends.
Expanded Vault Tooling
Finally, Euler is expected to improve the Euler Vault Kit (EVK) to make launching markets more accessible. Streamlined vault creation and expanded tooling for curators and DAOs have been discussed in forum proposals to lower barriers for protocols that want to design credit markets on Euler’s infrastructure.
Fluid
Fluid positions itself as the “next-generation liquidity layer” for DeFi. Built by Instadapp, it unifies lending, trading, and collateral management into a single capital pool. Instead of siloed markets where liquidity is fragmented, Fluid’s architecture allows assets to flow dynamically between lending vaults and its DEX, maximising capital efficiency. With innovations such as Smart Collateral, Smart Debt, and a liquidation system inspired by Uniswap v3, Fluid is seeking to redefine the design space for onchain credit.
Source: https://dune.com/queries/5011371/8288207
Fluid Market Size represents the Total Deposits in the protocol. The reason for choosing this metric over TVL is that debt is a productive asset in the protocol and contributes to the exchange’s liquidity.
Market Positioning
Fluid is the newest of the major money markets, but it has already reached an impressive scale. The protocol’s Market Size, defined as supplied liquidity, is over $3B across Ethereum, Arbitrum, Polygon, and Base, with Ethereum hosting the majority at $2.54B. Its TVL, which excludes borrowed assets, stands at around ~$1.4B, while more than $1.5B is actively borrowed.
Fluid’s integrated DEX reinforces this dynamic. In the past 30 days, it processed over $21B in trading volume, giving the protocol a dual identity as both a money market and a high-volume exchange. This hybrid structure distinguishes it from peers that separate lending from trading, and it is central to Fluid’s pitch as a unified Liquidity Layer.
Fluid has also expanded beyond EVM. In partnership with Jupiter Exchange, which has processed more than $995B in spot trading volume and operates one of Solana’s leading perpetual exchanges and liquid staking platforms, Fluid launched Jupiter Lend on Solana. Jupiter Lend mirrors Fluid’s lending architecture, with high LTV ratios and lower liquidation penalties. It has already reached a market size of over $1.3B and a TVL of more than $700M, and is currently the 12th largest lending protocol, signalling strong early adoption within Solana’s DeFi ecosystem.
Taken together, these metrics indicate that Fluid is carving out a distinct role. By combining lending and trading into a single liquidity engine and extending its Liquidity Layer to Solana through Jupiter, it is positioning itself as a capital-efficient liquidity layer that spans multiple ecosystems.
Core Features and Architectural Design
Fluid operates on a unified liquidity model, which sits at the centre of its architecture. Instead of separating lending, vaults, and trading into isolated silos, Fluid connects them through a single Liquidity Layer. Assets deposited here are recycled across the system, supplying capital to the Lending Protocol, Vault Protocol, and DEX Protocol. This structure increases capital efficiency while maintaining consistently high utilisation.
Lending Protocol
The Lending Protocol enables users to supply assets and earn yield from supply interest. Unlike traditional lending platforms, where deposits only back loans, here they become part of the shared liquidity that fuels the entire Fluid ecosystem. The system is also attractive for long-term lenders because it is designed to adapt to the advancements of the borrowing side without reallocating capital.
Vault Protocol
Fluid’s Vaults are single-asset, single-debt positions that offer some of the most aggressive parameters in DeFi. Loan-to-value ratios can reach up to 97% on certain pools, supported by Fluid’s Smart Debt and Smart Collateral mechanics. These vaults are designed for capital efficiency while still applying automated ceilings and risk controls to prevent systemic issues.
Smart Collateral allows collateral assets to earn both lending and trading fees, keeping them productive while securing loans.
Smart Debt allows borrowed assets to be reused instead of sitting idle as liabilities. In practice, this means borrowers can reduce their effective borrowing costs because trading fees from the pools they draw from are offset against their debt, all while retaining control of the borrowed capital and utilising it however they would like.
Liquidation Engine
Fluid employs a slot-based liquidation system inspired by Uniswap v3. Instead of liquidating an entire position when collateral falls, only the portion necessary to restore a healthy ratio is liquidated. Positions are grouped into ticks based on their loan-to-value (LTV) ratio, and liquidations occur within these ticks whenever the collateral value breaches the liquidation price.
The tick system is defined mathematically:
ratio = 1.0015^t, where t = tick
For example, consider a user who supplies $1,000 in ETH and borrows $800 in USDC. This results in an LTV ratio of 0.8. Using the equation, this position maps to a tick value of approximately -148.87. As the value of ETH falls, the tick value rises, and once it crosses a certain threshold, a portion of the position is liquidated.
Instead of wiping out the entire loan, Fluid liquidates only the portion required to bring the position back into a lower tick range. This targeted approach reduces liquidation penalties to as little as 0.1%, avoids unnecessary liquidations, and ensures that most of the position remains intact. The penalty is passed on as a discount to traders routing through DEX aggregators, making the process economically efficient for all sides.
As shown in the diagrams, only the affected ticks are liquidated, and the system quickly rebalances.
Fluid DEX
The Fluid DEX is integrated into the Liquidity Layer, turning collateral and debt into productive liquidity. Assets deposited by lenders or posted by borrowers also provide depth in trading pools, earning fees that further reduce borrowing costs.
DEX aggregators, such as KyberSwap and ParaSwap, among others, already route order flow through Fluid, providing it with external volume in addition to internal lending demand. In the past 30 days, the DEX has processed over $21B in swaps, highlighting its dual role as both a trading venue and a credit market.
The DEX is also evolving. DEX v1 was launched in October 2024, combining lending and trading into a single system. DEX v2, announced in April 2025, expands on this by allowing developers to deploy permissionless pools, experiment with custom fee models, and add new hooks. The aim is to turn Fluid into a modular playground for both traders and credit builders, while still anchored in its shared Liquidity Layer.
Future Roadmap
Fluid DEX v2 expands the protocol’s design space by introducing four types of DEXs, two of which are carried over from v1 and two that are entirely new. The existing types, Smart Collateral and Smart Debt, remain central to the system. They allow users to make both sides of a lending transaction productive, with collateral and debt positions contributing to liquidity and earning fees.
The innovation comes from the new Range Order types:
Smart Collateral Range Orders allow users to deposit collateral within a specific price range, similar to Uniswap v3. In addition to earning trading fees, the liquidity also accrues a lending APR and remains valid collateral. This creates a position that is both yield-bearing and credit-eligible.
Smart Debt Range Orders extend this logic to the borrowing side. Users can create range orders with borrowed assets, earning trading APR while simultaneously maintaining a debt position. This gives borrowers greater flexibility and reduces the effective cost of leverage.
Beyond these four types, Fluid v2 is designed to be extensible. Governance can approve additional DEX types, allowing the protocol to adapt as new trading primitives emerge.
DEX v2 also introduces several advanced features. Hooks enable developers to attach custom logic to pools, allowing for automation and specialised strategies in a manner similar to Uniswap v4. Flash accounting improves gas efficiency, making Fluid more attractive for CEX–DEX arbitrage and high-frequency trading. Finally, onchain yield-accruing limit orders ensure that even idle orders earn a lending APR while waiting to be filled.
Together, these innovations make Fluid DEX v2 more than just a trading venue. It is a programmable marketplace that blurs the lines between credit, liquidity provision, and trading strategy, and positions Fluid as one of the most experimental DEX–lending hybrids in DeFi.
Beyond this, Fluid is also working on upgrading Fluid to allow risk curators to deploy vaults and DEXs using the most advanced liquidation engine. In addition to this, the team is also developing a Fixed Rate Protocol (enabling users to lend and borrow on fixed terms), a Perps protocol, and sophisticated DEX LP vault strategies.
Conclusion
Onchain lending has matured into one of the most competitive and innovative verticals in DeFi, allowing users to gain additional exposure to assets or to unlock liquidity without selling their holdings. The vertical has continually evolved toward greater security, capital efficiency, and user experience, and today it stands as the second-largest sector after liquid staking.
There are more than 250 lending protocols live, but the market is highly concentrated. Aave alone represents nearly half of all value locked in the category, while the following nine protocols combined account for just under 40%. This concentration reflects both the resilience of incumbents and the challenge for new entrants to carve out a share in a competitive landscape.
The four protocols examined here highlight the range of design choices that will shape the markets’ future. Aave dominates through scale and conservative risk, making it the trusted venue for institutions and retail alike. Morpho optimises existing liquidity, closing inefficiency gaps while building new rails for peer-to-peer credit. Euler pursues programmability, providing builders and DAOs with infrastructure to design custom credit markets. Fluid pushes for unification, collapsing the boundaries between lending, vaults, and trading into a single liquidity layer that spans both EVM and Solana.
What emerges is not a single model for onchain credit but a spectrum. Pooled markets, optimiser layers, programmable vault systems, and unified liquidity architectures each address different needs. Together, they reflect a sector moving beyond simple borrow-and-lend mechanics toward platforms that manage liquidity, risk, and yield as integrated systems.
As lending enters its next growth wave, this diversity of approaches is a strength for the ecosystem. It ensures that users, treasuries, and institutions can choose the model that best fits their needs, rather than relying on a single blueprint.
The next part of this series will transition from architecture and positioning to risk frameworks and capital efficiency, offering a deeper examination of how these protocols behave in real-time and how they manage stress across various market conditions.
Glossary
Market Size (Supplied Liquidity): Market Size is the value supplied to the protocol. It includes all collateral and deposits, regardless of whether those assets are currently utilised for borrowing or not.
TVL (Total Value Locked): Total Value Locked in the protocol smart contract. TVL excludes the amount borrowed from the protocol. In other words, TVL = Supplied Liquidity - Borrowed Assets.
LTV (Loan-To-Value): The LTV ratio defines the amount of funds that can be borrowed by depositing collateral. An LTV value of 85% means that, on a deposit of $1,000, a user can borrow up to $850. LTV = Borrowed Capital Value / Collateral Value
Liquidation Threshold: Liquidation Threshold is higher than LTV and is the point at which liquidation can occur. A position could reach its liquidation threshold when the value of the collateral is dropping and/or there is interest accrued on the position.
APR (Annual Percentage Return): APR represents the interest rate earned on deposits, expressed as a percentage, without accounting for compounding.
HF (Health Factor): A position health factor measures how safe a borrower’s position is. It is the ratio between the value of supplied collateral (adjusted by liquidation threshold) and the value of borrowed assets.
HF = Collateral Value * Liquidation Threshold / Borrowed Assets Value
HF > 1 means the position is safe
HF = 1 means the position is at the liquidation threshold
HF < 1 means positions can be liquidated
 | A guest post by
|
 | A guest post by
|
gg
Fluid seems very undervalued