TL;DR Intents are changing DeFi and users are getting better onchain experiences as intent-based systems grow. Intents are best known for offering end users speed and price benefits but they also offer token issuers (and other LPs) a way to provide liquidity in an MEV-aware manner. Arrakis proves this with HOT AMM, which empowers LPs through an MEV-aware, intent-based design.

Key Takeaways:

• Intents offer meaningful UX benefits and they’re growing fast as DeFi users seek out seamless onchain experiences.

• Solvers aim to fulfill users’ intents when the incentives align to do so. Solvers are typically sophisticated actors because the entry barrier is high.

• Centralized entities have begun to dominate major token pairs, leaving LPs more exposed to toxic flow.

• A new class of intent-based systems offer token issuers a way to provide liquidity in a sophisticated, MEV-aware manner.

• ERC-7683 is the first crosschain intents standard and has quickly gained adoption this year. It uses intents to unify Ethereum.

• DeFi needs to protect LPs and support longtail assets to build sustainable markets. Arrakis is taking an MEV-aware approach to achieve this goal with HOT AMM.

DeFi is evolving fast and forward-thinking builders are laying the foundations for the future. But today’s most promising new solutions frequently come with trade-offs. Ethereum’s L2 era is evidence of this, as recent scalability improvements come at a cost of fragmentation.

The rise in intents has ushered in one of the biggest changes in DeFi in recent years. Intent-based systems offer significant UX improvements by asking third-party solvers to commit actions on behalf of users. Although the possibilities are endless, in practice so far intents have primarily been used for crosschain bridging or swapping.

To date, the intents boom has benefited regular users and a small number of centralized market makers. But as we uncover in this piece, cutting-edge systems like HOT AMM are using intents and MEV awareness to protect token issuers and LPs. We explain how this model could disrupt the DeFi landscape in depth below.

What Are Intents?

Put simply, intents let users specify their desired end-state or outcome for a transaction. Paradigm summarizes intents as follows:

In the standard transaction-based flow, a transaction signature permits the validator to follow exactly one computational path against a certain state while a tip incentivises the validator to do so. On the other hand, an intent does not specify exactly what computational path must be taken, but rather allows for any which satisfy certain constraints. By signing and sharing an intent, a user is effectively granting permission to recipients to choose a computational path on their behalf. This distinction allows a slightly more rigorous definition of intents as signed messages which allow for a set of state transitions from a given starting state, a special case of which is a transaction which allows for a unique transition. That being said, we will continue to refer to “intents” as distinct from transactions.

In other words, intents are essentially a shift from following a specific step-by-step route onchain to derive an outcome to declaring an outcome and letting third-party actors (solvers) find the best route to arrive at your outcome.

So far, intents have changed the way we think about swapping and bridging. With intents, users can go to an AMM and offload the routing to solvers or they can go to a bridge and ask to receive a different token at their requested destination. Early adopters in this space include CoW Swap, UniswapX, and Across. CoW Swap and UniswapX rely on a network of solvers to offer users MEV-protected token swaps while Across enables seamless crosschain bridging (Uniswap also integrates Across for in-app crosschain transfers).

Why Intents Are the Future

Intents have gained rapid adoption since 2023—with intent-based systems now accounting for about 30% of DeFi volumes—and there is a reason for that: Users want fast, cheap outcomes and intents offer both.

Intents offer users other benefits too. As solvers compete to fill orders, users achieve their desired outcome fast while the solver must wait for settlement. This also means intents offload risk onto the solver, which is another win for users.

When a third party is responsible for filling the order, swappers are protected from sandwich attacks and crosschain users are not exposed to finality risk.

In short, intents offer users superb onchain experiences and that’s why they’re growing. As users will increasingly demand good UX in the future, this trend is set to continue.

Why Solvers Fulfill Intents

Solvers have a profit margin cooked into every quote they offer to a user. They may also earn extra incentives, as is the case on CoW Swap, where solvers that win the batch earn COW rewards. Moreover, as they are sophisticated actors, they often have access to information that the user does not have, such as the best price on external markets. So while the user gets a good price with MEV protection when they ask for a swap, the solver still receives a small cut for filling the order in the same way a market maker or CEX would.

Solvers are in the business of securing profits and they make a significant outlay to participate in intent markets. First, they must spend on infrastructure that gives them a competitive edge over other solvers. They may also have to put effort into sourcing inventory and offering users correct prices, a burden Tarun Chitra et al. describe as “congestion costs” in An Analysis of Intent-Based Markets. Solvers are incentivized to profit so they pay these costs.

Solvers use different strategies to compete in intent markets. For example, some are operated by sophisticated actors like Wintermute. These firms also engage in strategies like searching and block building, vertically integrating into the MEV supply chain to extract profits. Others like Copium Capital, an elusive one-man entity that regularly dominates CoW Swap’s order flows, win significant volumes solely by outcompeting other solvers on complex routing to AMMs (without relying on any private liquidity).

But all of the top entities share the same characteristics: they are sophisticated, well-capitalized organizations. As the costs associated with fulfilling intents is high, the entry barrier is high and few can compete with the top players.

How Private Market Makers Dominate Intents

We know that users go to intent protocols because they want good UX and solvers fulfill users’ intents because they want to profit.

But as the entry barrier is high for solvers, private market makers (PMMs) like Wintermute currently dominate intent-driven order flows. We refer to entities like Wintermute as Centralized Vertically-Integrated Market Makers (CVMMs) because they operate across the MEV supply chain to profit off retail.

Protocols like UniswapX see the bulk of their volume captured by Wintermute while some smaller systems rely on their own solvers, which is also problematic. So while intents have improved DeFi’s UX and end users are better off, PMMs dominate intent-driven order flows today.

How the Intents Boom Impacts LPs

We’ve established that end users and solvers are incentivized to participate in intent markets. But the intents boom impacts another actor: LPs.

Searchers extract MEV from DeFi every day, with most leaked through arbitrage. This has created a cost to LPing known as Loss-Versus-Rebalancing. LPs primarily aim to offset this cost by earning fees for every swap they facilitate.

Users can execute a swap through a traditional AMM or intent-based system. On traditional AMMs, LPs offer stale prices because CEXs update faster than DEXs and this is how sophisticated arbitrageurs extract value. On intent-based systems, PMMs dominate flows because the market is competitive.

As PMMs dominate flows through intent-based systems, they leave LPs more exposed to toxic flow. In other words, while PMMs capture intent-driven flows, arbitrageurs are eating the LPs’ lunch. We think it’s time to leverage the possibilities of intents to fix this.

Major Pairs and the Longtail

While firms like Wintermute can comfortably fulfill requests on pairs like ETH/USDC, catering to the longtail is much harder.

CVMMs hold major tokens and they may also borrow assets from a token issuer to market make on CEXs. They can hedge their risk on major pairs but this isn’t possible for every longtail asset.

As CVMMs are restricted by their inventories, LPs currently dominate volumes for longtail assets. This also means that intents don’t typically service the longtail today.

This directly impacts users: While they can expect a seamless experience when they request a swap from ETH to USDC, they don’t get the same capital efficiency on less liquid pairs because they’re not serviced through intent-based systems.

While intents are growing, longtail assets remain underserved. This will become a more pertinent issue in a world of millions of tokens.

How ERC-7683 Is Unifying Ethereum

Although intents do not service the longtail, we can confidently conclude that intents will shape the onchain future because of the superior UX offered on major pairs alone.

We also know that intents are growing in popularity. While volumes on intent-based systems continue to rise, intents are gaining adoption as a way to improve Ethereum UX. The growing support for the ERC-7683 standard co-authored by Across and Uniswap Labs proves this. ERC-7683 was developed in response to Ethereum’s fast-growing L2 landscape and it establishes a unified framework for crosschain intent-based systems.

Optimism adopted ERC-7683 for the Superchain, other L2s like Polygon and Taiko are also early supporters, and key Ethereum voices like Vitalik have publicly endorsed it throughout this year. This standard’s biggest proponents argue that intents will unify Ethereum.

Arrakis supports ERC-7683 because we believe that intents will power the interoperable future. Liquidity fragmentation is a major problem for L2 explorers and token issuers and crosschain intents offer a solution.

The likes of CoW Swap and Across pioneered intent-based architectures to offer users better onchain experiences. Now, ERC-7683 aims to use intents to make Ethereum more interconnected.

How Arrakis Fits Into the Intent-Based Future

In today’s DeFi landscape, the intents boom offers excellent UX to users and healthy volumes to PMMs. But longtail assets are underserved and LPs could be better supported.

Intents will be part of Ethereum’s future. But we need a system that offers the following:

• Good UX and prices for users.

• Protection for LPs.

HOT AMM, the MEV-aware DEX Arrakis built on Valantis, answers to this need. With HOT, Arrakis runs a Request-for-Quote system, which uses an offchain component to issue quotes to solvers. The quotes are determined by an offchain model running an Alexander & Stoikov strategy that is much more information-aware than a traditional AMM because it ingests pricing from CEXs and is aware of pricing on other onchain venues. When solvers  trade against the pool, LPs receive healthy flow because all the solver volume comes from retail traders on CoW Swap, UniswapX or 1inch Fusion. Furthermore, for the permissionless onchain side of the DEX, HOT adopts a dynamic fee model, which eliminates arbitrage opportunities. Through HOT’s MEV-resistant, intent-centric design, Arrakis offers a solution that uses intents and MEV awareness to empower LPs. In doing so, we hope to take on the extractive actors that monopolize onchain markets today.

HOT already accounts for 20%+ of CoWSwap’s daily ETH-USDC volumes on mainnet. In the future, we could use intents to serve LPs flow from multiple chains. With ERC-7683 unifying L2 liquidity, we have a way for LPs to provide liquidity on one chain and receive intent-driven flow from another. Arrakis is building towards this future and we anticipate that solutions like Across (for cross-chain bridging) and UniswapX (for cross-chain swapping) will play a key role in helping LPs receive crosschain intent-driven flow.

We recently launched HOT as part of our plan to vertically integrate into the MEV supply chain. Token issuers can use HOT to provide liquidity in a more sophisticated manner, benefiting from MEV protection and improved returns. Moreover, a similar setup to HOT will be available on Uniswap V4 once it launches.

As a decentralized alternative to CVMMs like Wintermute, Arrakis is also well positioned to scale to support longtail assets in the future because it is not restricted by a limited inventory. To learn more about HOT, check our full announcement here and explore the app here.

References

• An Analysis of Intent-Based Markets [Tarun Chitra et al. in arXiv:2403.02525v2]

• Are Retail LPs Going Extinct? Exploring the MEV Supply Chain [@dreamsofdefi for Arrakis Finance]

• CoW Documentation [CoW]

• ERC-7683 Documentation [Across and Uniswap Labs]

• ERC-7683 Under the Hood [@Uniswap via X]

• How intents can solve multi-chain DeFi composability [@dreamsofdefi for UMA]

• HOT, the MEV-Aware AMM Built to Empower LPs, Is Live [@dreamsofdefi for Arrakis Finance]

• Hybrid Order Type: A New MEV Aware AMM Design [Arrakis Finance and Valantis Labs]

• Intent-Based Architecture and Their Risks [Georgios Konstantopoulos and Quintus Kilbourn for Paradigm]

• Introducing the Arrakis DVMM: DeFi’s First MEV-Aware Onchain Market Maker [@dreamsofdefi for Arrakis Finance]

• LVR from Theory to Practice: A Survey of Next Generation DEX Designs [Ari Rodriguez and Daniel Contreras for Arrakis Finance]

• Optimism leverages intents as Across powers Superchain bridge interface transfers [@dreamsofdefi for Across]

• Solverscan Data [Solverscan]

• The AMM Renaissance: How Offchain Systems Reduce MEV and LVR [@dreamsofdefi for Arrakis Finance]

• The case for canonical asset bridging: Secure value transfer and cross-chain interoperability [@dreamsofdefi for Across]

• Uniswap Labs Integrates Across for In-App Bridging [Across]

• Vitalik Buterin discusses crosschain experiences [@AcrossProtocol via X]

• Why We Need to Fix DeFi: Addressing the CVMM Problem [@dreamsofdefi for Arrakis Finance]