Aave V4: Reshaping the underlying architecture of Decentralized Finance lending protocol

As one of the core components of the DeFi ecosystem, any move by Aave, the largest and most mature lending protocol, attracts significant attention from the industry. Recently, at the highly anticipated ETHCC conference, Aave's founder officially announced that the team is about to launch its next major iteration version - Aave V4.

Aave V4 is not just a simple routine upgrade, but a key milestone in Aave's long-term strategic plan for 2030. This upgrade was first proposed officially in May 2024, with the core goal of systematically addressing the limitations exposed by the V3 version during its operation, particularly making breakthroughs in critical areas such as scalability and risk management. Through this far-reaching update, Aave aims to fundamentally reshape the underlying architecture and core functions of the Decentralized Finance lending protocol, preparing for the future development of the protocol.

In this article, we will explore in detail what Aave V4 contains. We will review its evolution, analyze its new architecture, and interpret these changes within the broader trends of the Decentralized Finance industry.

The Evolution of Aave

Aave's journey began with ETHLend, a P2P platform where lenders and borrowers needed to find each other's counterparties, but the process of finding matching counterparts was slow and fraught with uncertainty. After deeply recognizing these fundamental flaws, the team upgraded the brand from ETHLend to Aave (i.e., AAVE V1) in September 2018, decisively shifting from a P2P model to a liquidity pool-based Point-to-Contract (P2C) model, where funds were pooled together to enable instant lending. The subsequent Aave V2 further reduced transaction costs on the congested Ethereum network by optimizing smart contracts, thereby allowing more people to access Decentralized Finance.

The current version Aave V3 has made significant strides in capital efficiency and risk management compared to version V2. It introduces several key features, such as:

• Efficient Mode (E-Mode): When the prices of the assets deposited and borrowed by users are highly correlated (for example, between stablecoins, or between ETH and stETH), E-Mode allows users to unlock higher borrowing capacity (such as a higher LTV). This directly addresses the issue of insufficient capital efficiency for correlated assets in V2.

• Isolation Mode (Isolation Mode): Allows new, higher-risk assets to be launched in an "isolated" manner. The collateral provided under isolation mode can only be used to borrow a set of governance-approved stablecoins, with a clear debt limit, and cannot be mixed with other collateral. This effectively "isolates" the risk of new assets, preventing risk contagion.

However, Aave V3 also exposes a deeper strategic limitation: a single entity architecture cannot flexibly respond to the demands of emerging markets and diversified scenarios. Imagine a traditional bank that initially only accepts real estate as collateral. All its forms, processes, and risk assessment models are designed around real estate. Now, a client wants to apply for a loan using their company’s equity, patents, or even future accounts receivable. The bank will find that its existing "one-size-fits-all" process is completely inadequate for handling these new types of assets with different risk characteristics. The bank either has to undergo a painful internal reform or abandon these new businesses.

Aave V3 faces a similar dilemma. Its core smart contracts are tailored for crypto-native assets (such as ETH, WBTC, and stablecoins). When the industry began to introduce RWAs—such as tokenized government bonds or private credit—as collateral, Aave V3's single architecture proved inadequate. RWAs involve off-chain legal compliance, counterparty risk, and different liquidation logic, which cannot simply be integrated into the existing smart contract framework.

This is the core issue that Aave V4 aims to fundamentally address: how to evolve from a single rigid product into a flexible platform capable of supporting countless financial scenarios.

Aave V4: Modular New Architecture

Aave V4 introduces a brand new design called the "Liquidity Hub + Spoke" model. This architecture is a direct response to the limitations of a "single entity" and can be understood through a simple analogy from traditional finance: a central bank and its network of commercial banks.

• Liquidity Center: Aave's "central bank"

  • On every blockchain network running Aave, there is a unified Liquidity Hub that aggregates all assets supplied by users. This hub serves as the central liquidity source for the entire network. It does not directly provide "retail" services to end users. Instead, it focuses on macro liquidity management and risk control, providing stable and deep liquidity for the entire ecosystem. This model is expected to improve capital utilization, bring higher returns to lenders, and offer lower interest rates to borrowers.

• Liquidity centers on different chains are not isolated islands; rather, they can efficiently communicate and transfer liquidity with each other. This is mainly achieved through a mechanism known as the "Unified Cross-Chain Liquidity Layer" (CCLL), which is fundamentally supported by Chainlink's Cross-Chain Interoperability Protocol (CCIP).

• Spoke: Aave's "specialized commercial bank". The liquidity center operates in the background, and users interact with the protocol through various Spokes. Spokes are user-facing, modular lending markets, each designed for a specific purpose and connected to a central liquidity center. They function like specialized commercial banks. For example, there might be:

  • Core Spoke: A general lending solution for handling blue-chip crypto assets with low risk and high liquidity, such as ETH and WBTC.

  • E-Mode Spoke: Specifically optimized for stablecoins, LSTs, and other highly correlated currency pairs, providing the highest capital efficiency.

• RWA Spoke: Tailored for tokenized treasury bonds, real estate, and other real-world assets. This type of Spoke can integrate stricter access, custody, or compliance rules to meet institutional and regulatory needs.

  • A high-leverage trading Spoke designed for professional traders seeking high-risk, high-reward opportunities, featuring a special interest rate model and risk management parameters.

The most important aspect of this design is its openness. Aave V4 will allow developers to build and propose their own Spoke. If a new Spoke design is approved by Aave's governance, it can receive a line of credit from the liquidity hub, thereby leveraging Aave's vast liquidity network to launch a new, specialized market. This fundamentally transforms Aave from a mere product into a foundational platform for financial innovation.

Comparison: Aave VS. Sky (formerly MakerDAO)

In order to fully understand Aave's strategic direction, it would be helpful to compare it with its main competitor, Sky. Sky has recently undergone a rebranding and launched its own "Endgame" plan. It can be said that "great minds think alike"; Sky has also adopted a modular architecture, marking a shift across the industry towards a more flexible and scalable design.

similar

The architecture of Sky can be described as "Sky Core + SubDAO".

• Sky Core plays the role of a "central bank" in the Sky ecosystem, inheriting the function of issuing stablecoins (now USDS, originally DAI). It establishes the most core rules (for example: which SubDAOs are allowed to access the system, the total minting limit for each SubDAO, emergency shutdown mechanisms, etc.), maintains the stability of USDS, and serves as the ultimate credit and security guarantee.

• SubDAO is a semi-independent specialized organization operating within the Sky ecosystem, acting as a "commercial bank" for specific domains. The core tasks of SubDAO are asset management and risk assessment. They are authorized by Sky Protocol to receive specific types of collateral and initiate a request to mint USDS to Sky Core. For example, Spark Protocol is currently the only mature SubDAO in the Sky ecosystem, focused on lending, and is a direct competitor of Aave. Other SubDAOs may focus on RWA assets or other niche markets.

The similarities between Aave's "Liquidity Hub + Spoke" and Sky's "Sky Core + SubDAO" are evident: both recognize that a single entity cannot meet all market demands, and therefore adopt the model of "central bank + specialized commercial banks": the central commercial bank formulates policies and provides liquidity, while specialized commercial banks are responsible for developing specific business scenarios.

Looking back at the grievances between the two projects, Sky Spark was born by directly forking the open-source code of Aave V3. The two sides also had a fierce dispute over the profit-sharing protocol, with Aave accusing Spark of failing to pay the promised 10% profit share. Now, Aave V4 has merely "borrowed" Sky's mature modular design concept, which can be seen as "using their own methods to return to them."

different

Despite their similarities, Aave and Sky also have significant differences in core business, economic models, and ecological sovereignty.

First, let's talk about the types of liquidity: Aave's Liquidity Hub aims to provide liquidity for a wide range of asset classes, including stablecoins, volatile assets (such as ETH), and derivative assets (LSTs). Sky inherits the original genes, with its core strategy always revolving around the issuance, stability, and promotion of its native stablecoin USDS (formerly DAI). The main task of its SubDAO is to create more application scenarios and demand for USDS, deepening its liquidity moat.

Secondly, there is the economic model and sovereignty: this is the most fundamental difference between the two. The Sky SubDAO is granted a high degree of economic sovereignty, with each SubDAO allowed to issue its own governance token (such as Spark's SPK token), enabling it to build independent economic models, implement its own incentive programs, and directly capture the value created by its own business growth. This economic independence allows SubDAOs to evolve complex and powerful functional architectures. Taking Spark, the only mature example in the current Sky ecosystem, as an example, its operational model can be likened to a dual-layer financial system:

1. "Commercial Bank" level ( retail end ): It has a lending platform for end users called Spark Lend. This part of the business directly serves individual users, functioning similarly to the commercial banks we are familiar with.

2. At the level of "Regional Reserve Banks" ( wholesale end ): Spark also has a liquidity layer called Spark Liquidity Layer (SLL), which acts as a regional "liquidity hub". After obtaining liquidity (such as USDC/USDS) from Sky Core, SLL not only provides funding support for its own "commercial bank" Spark Lend, but also "wholesales" this liquidity to other Decentralized Finance protocols, such as Morpho, and even competitors like Aave.

Therefore, Spark is not a simple lending application, but a liquidity engine that integrates retail and wholesale business, fully utilizing its SubDAO identity to create and allocate value both within and outside the Sky ecosystem.

In contrast, the independence and autonomy of Spokes in Aave V4 are much weaker. Currently, Spokes cannot issue their own tokens. They are extensions of the core Aave protocol, and the value they generate (such as interest income) will flow back to the Aave DAO. A Spoke is similar to different divisions under a large corporation, operating under the unified Aave brand and economic framework, with the value created also flowing back to the corporate headquarters.

Macro Perspective

The architectural changes of Aave and Sky are not isolated events, but rather a direct response to the major trends shaping the future of Decentralized Finance.

Integrate RWA

The next frontier of DeFi growth is widely believed to be the tokenization of real-world assets such as government bonds, real estate, and private credit. These assets come with unique legal and compliance requirements, making it difficult to manage them within a single, large protocol. The modular architecture of Aave V4 and Sky is highly suitable for this, as it allows protocols to create independent, customizable, and even permissioned "sandbox" environments (such as RWA Spoke or RWA SubDAO) while maintaining their core decentralized and permissionless characteristics, specifically designed to hold and manage RWA.

application chain's