Author: @Cryptovoxam

Translation: Blockchain in Vernacular

Intent-driven applications will shape decentralized finance (DeFi) 3.0, and if you haven’t realized this yet, it’s probably because you haven’t understood the potential that intent can unlock.

I will write this article to tell you about the decentralization intention:

It would be impossible to explore every possible use case that intents unlock, as that would require an infinite number of threads. I wanted to keep this as compact as possible.

I want to focus on a specific area of ​​finance where trillions, some estimates even run into quadrillions of dollars, are moved around every year in the traditional financial system.

Before we dive into the future of on-chain derivatives, let’s take a look at current models and their main trade-offs.

The following is a common classification:

1. CLOB (Central Limit Order Matching) Model

This is the model that every centralized exchange like Binance is using and the first DeFi implementation was done by @dYdX.

The reason every trading platform uses an order book is because it is the best infrastructure model they can use. But if that is true, why did dozens of other teams full of smart people decide to go another way?

This is because order books require sophisticated market makers to actively provide liquidity and fast order matching. While the former is easier to achieve, the latter is sometimes impossible. Imagine building an order book on the Ethereum mainnet with a 12-second block time.

This is why many teams have decided to move their matching engines off-chain. dYdX V3, Aevo, RabbitX, etc. are all great examples, but their amazing speed comes at the cost of decentralization.

Some projects have successfully built fully on-chain order books using alternative virtual machines (altVMs). The best examples are Hyperliquid, which I really like, and the V4 version of the giant dYdX.

2. Liquidity Provider (LP)-based Model

This is a large category that encompasses several sub-models with subtle differences between them. A common feature is that price discovery is done outside the protocol. They use oracle providers similar to @PythNetwork and @chainlink or custom price feeds.

This would be the worst case scenario because not only would you be affected by the fall in the asset price, but you would also have to pay out profits to traders. Your capital would be destroyed.

However, there are some advantages.

Since they use oracles to price assets, you can achieve zero-slippage trading, which can be very interesting for traders, especially for long-tail assets, and that’s not all. As a DeFi Maxi, one of the features I like about DeFi is its composability.

Tokens like @GMX_IO's GLP or @JupiterExchange's JLP are composable. You can use them as collateral in loans, trading, or in certain leverage strategies. These use cases don't exist in other perpetual decentralized exchange models.

3. AMM (Automated Market Maker), vAMM (Improved Automated Market Maker) and Hybrid Models

While perpetual contracts like @DriftProtocol and @perpprotocol V1 use AMM and vAMM structures, they are now considered outdated models.

Interestingly, they are now being used in hybrid models.

@vertex_protocol has a price/time algorithm: orders will be executed at the best available price, whether that's from an order book or an AMM.

@DriftProtocol is similar, but it even has a third source of liquidity: they call it JIT liquidity.

JIT stands for Just-In-Time, which is a Dutch auction model.

This approach is very interesting because it is similar to the mechanism used by intent-driven protocols. For example, UniswapX and 1Inch Fusion utilize a Dutch auction model in order to let solvers efficiently satisfy intents.

4. Aggregator

They aggregate orders from multiple decentralized exchanges (dexes) and offer the best price across all the consolidated trading venues. They are also able to split trades across multiple platforms.

They will usually also have their own pool of funds.

Among the aggregators, there is also @vooi_io who are developing a cross-chain aggregator (EVM + AltVM).

4) Solver Model (also known as Intent Driven)

Broadly speaking, we view a solver (aka ticket filler/relay) as an off-chain agent with an economic incentive to satisfy the user’s intent.

In perpetual contracts, the solver is the market maker that takes the opposite side of you.

In the derivatives world, solver implementation is still very immature. However, these models have seen significant adoption when it comes to the rest of the cryptocurrency space.

You can see their growth below:

One of the pioneers of this model is @DriftProtocol, whose V2 version went live at the end of 2022 and introduced the JIT liquidity mentioned above.

Another player in the space is @symm_io, which allows bilateral agreements (RFQs) between two parties: traders and solvers.

In this context, the solver is also known as a “hedger.” Market makers generally do not take price risk: if they take an opposite position to yours, they will need to hedge that trade elsewhere.

The interesting concept here is that on-chain users are transacting with off-chain liquidity.

Symmio focuses on building the backend and infrastructure of a sustainable decentralized exchange platform that third-party teams can leverage for development.

@CadenceProtocol is also building a similar system.

@UniDexFinance is a perpetual contract aggregator built on @MoltenL3, introducing the PrMM (Programmable Market Maker) funding pool.

This is an interesting concept as it allows for the creation of programmable pools of capital that are fully customizable to run specific market making strategies.

5. Summary

The intent-driven space is really fascinating and could really become a foundational element for the next generation of decentralized applications. Despite its strong value proposition, the space is still in its early stages. There are three main challenges:

1) Solver competition leads to centralization.

2) Fragile solver infrastructure for complex intents.

3) The threshold for deploying and operating solvers is high.

But one of the main reasons I’m confident is that some of the smartest people are working on these problems. People like: @EverclearOrg; @EnsoFinance; @aori_io; @anoma; @intentessential; @ApertureFinance, etc.

As far as the derivatives DEX space is concerned, I think it is at the stage of undergoing a revolution in intent:

1) What traders need most is speed and liquidity.

2) Since they are transacting on-chain, they also care about permissionlessness and self-custody.

3) A mature intent-driven domain can meet all of these needs.