What if the #1 lending protocol on Sui launched a DEX?

Key Takeaways

• While TVL remains one of the most widely used metrics in the DeFi ecosystem, it has fundamental limitations—being highly sensitive to token price fluctuations and failing to reflect the actual utility and availability of liquidity. As a result, the industry is beginning to recognize the need for a paradigm shift—away from simply measuring locked value, and toward evaluating how effectively liquidity can be utilized and connected across multiple protocols, i.e., through the lenses of liquidity availability and composability.

  
  

• From this perspective, Suilend and STEAMM stand out not as two isolated protocols, but as a highly integrated DeFi stack built on Sui, operating in close synergy with one another. Through innovations such as superfluid AMMs, a modular Quoter architecture, and seamless liquidity routing, the two protocols leverage both the unique infrastructure of Sui and a deeply considered design philosophy to redefine the standards for liquidity efficiency and user experience.
  

• At the heart of this strategy lies a team with a proven track record from the Solana ecosystem, whose consistent execution and deep technical understanding of Sui enable them to build a synergy structure that is increasingly difficult for others to replicate. With products like Rootlet, SpringSui, and Attention forming part of a coherent, long-term strategy, this team is well positioned to establish the most robust and self-reinforcing DeFi infrastructure within the Sui ecosystem.

TVL (Total Value Locked) is one of the most widely used metrics in the blockchain industry, representing how much liquidity is locked within a given network. It is commonly referenced to compare the scale of DeFi ecosystems across chains or to gauge the level of trust a particular protocol has earned in the market.

However, despite its popularity, TVL as a metric has several critical limitations. First, because TVL is denominated in USD, any price increase in the native token of a blockchain can inflate the metric—even if no new liquidity has actually entered the ecosystem. In such cases, the metric may give a distorted picture of growth. Second, when liquidity is heavily concentrated in a specific protocol—such as a lending platform or a DEX—it cannot be freely utilized across the broader ecosystem. Liquidity locked in one protocol is generally siloed and not readily available for use elsewhere.

In other words, TVL has inherent structural shortcomings in measuring the actual availability of liquidity on a blockchain. A high TVL figure alone does not guarantee a healthy DeFi ecosystem. As DeFi continues to mature, there is a growing recognition that the focus should shift from TVL to metrics that reflect the efficiency and composability of liquidity.

This perspective aligns with what I discussed in my previous research on liquidity availability:

In this context, liquidity availability focuses not on the simple amount of locked-up funds, but on how much liquidity is actually available for any types of transactions, and how quickly and stably that liquidity can be utilized in any situation.

Therefore, the key to a sustainable and mature DeFi ecosystem lies not in artificially inflating TVL, but in efficiently utilizing idle assets locked within protocols. For TVL to function as genuine infrastructure liquidity, decentralized exchanges must be able to tap into the idle assets sitting in lending protocols—or vice versa. Unfortunately, this level of interoperability is rare in practice, as few smart contract platforms allow real-time, low-latency interaction between independent contracts. Furthermore, when DEXs and lending protocols are operated by separate stakeholders, liquidity sharing introduces significant governance and security risks.

But what if we imagine a composable blockchain environment that enables atomic-level interoperability between smart contracts, and within that ecosystem, a team that has already proven its ability to launch a successful lending protocol builds a new decentralized exchange? Provided the team has sufficient technical and operational capability, this could represent one of the most compelling realizations of the “ideal DeFi environment” I described earlier.

This brings us to a real-world example: the Sui ecosystem, where we are witnessing a live experiment in liquidity efficiency—Suilend and STEAMM. Suilend is currently the largest lending protocol on Sui, and the same team recently launched STEAMM, a decentralized exchange that now allows them to actively shape both pillars of DeFi: lending and trading. In this article, I explore how these two products create synergy when operated in tandem, how they optimize liquidity availability, and finally, why the Sui blockchain offers an ideal foundation for realizing such a tightly integrated DeFi stack.

1. Suilend: How a Slow Starter Became the Market Leader

Source: DeFiLlamma

While Suilend is now the largest lending protocol in the Sui ecosystem, that wasn’t always the case. Back when Sui first launched its mainnet, few had even heard of Suilend. In the early days, the ecosystem was dominated by two other lending protocols—NAVI and Scallop—both of which had already established themselves by the time Suilend entered the scene.

And yet, despite its late arrival, Suilend rapidly gained traction, steadily capturing TVL and market share to become the undisputed leader in Sui DeFi today. It was a slow starter, but one that ended up outpacing everyone else.

So what exactly allowed Suilend to rise from obscurity and take control of the ecosystem so quickly? Let’s take a closer look.

1.1 An Experienced Team and Founder: From Solana to Sui

Source: Solana Floor

When Suilend first entered the Sui ecosystem, what immediately caught the attention of many was the experience of its team and founder. Prior to launching on Sui, the team had already built a successful lending protocol on one of the most prominent non-EVM chains—Solana. That protocol was Save (formerly known as Solend), which at its peak during Solana’s DeFi summer surpassed $1 billion in TVL and ranked as the number one DeFi protocol in the ecosystem. This track record alone was enough to position Suilend as a project worthy of trust. But beyond just having been “number one” before, what truly mattered was that a team that had built and operated a product through an entire ecosystem cycle was now launching a similar product in a new environment. That continuity and operational maturity carried weight.

The team’s experience on Solana was also expected to translate into real advantages on Sui. In particular, Solana’s use of Rust and Sui’s adoption of the Move language share similar design philosophies, especially around safety and low-level performance. This led to the perception that a team deeply familiar with Rust-based systems would be well-equipped to quickly adapt to Move, and deliver a sophisticated, production-ready lending protocol on Sui. This technical alignment added to the credibility and excitement surrounding Suilend’s launch.

1.2 Rooter and the Rise of the Rootlets: A Personal Branding Case Study

Source: Rootlets

Another key factor behind Suilend’s rise is its founder—Rooter. Among all the founders in the Sui ecosystem, Rooter stands out as someone who has invested heavily in building his personal brand. He is one of the most active founders on Twitter, frequently sharing insights that only someone with his background could offer—drawing from firsthand experience developing protocols on both Solana and Sui. His content has ranged from technical comparisons between the two chains’ development environments to detailed explanations of why he chose Sui as the next frontier after Solana. Notably, at the second Sui Basecamp in 2025, he spoke publicly about the differences in tooling and developer experience between the two ecosystems, effectively positioning himself as a public ambassador for Sui.

These posts not only resonated with builders and users within the Sui ecosystem but also attracted the attention of developers and investors from other chains. As a result, Rooter’s personal visibility helped build organic trust and curiosity around the Suilend protocol. In a space where attention is one of the most valuable resources, Rooter’s strategic use of personal branding became a meaningful advantage—one that helped amplify Suilend’s reach and reputation far beyond its initial footprint.

Source: Rootlets

Another noteworthy initiative led by Rooter is the NFT project “Rootlets”, which has gained significant attention from a branding perspective. Although it initially faced criticism within the community due to its relatively high mint price, the narrative quickly shifted. Starting with airdrops from prominent ecosystem tokens—such as the memecoin $FUD, widely recognized within the Sui ecosystem—Rootlets holders began receiving various token rewards. As a result, the floor price of Rootlets NFTs surged past 200 SUI, positioning it as a serious contender to Prime Machin, the leading NFT collection in the ecosystem.

What’s even more remarkable is that this momentum occurred before Rootlets holders received a substantial airdrop of $SEND, the governance token of Suilend. This helped completely silence the earlier complaints about Rootlets’ pricing, as market sentiment turned overwhelmingly positive.

One particularly interesting aspect of Rootlets is how Rooter used the project as a vehicle to draw influential figures from other ecosystems into the Sui ecosystem. While it’s ultimately a subjective matter whether simply holding a Rootlets NFT means someone has truly entered the Sui ecosystem, the symbolic impact was clear.

Rootlets created custom “Honorary Rootlets” for select high-profile individuals and gifted them as a gesture of recognition. Notably, this included Mert, a leading influencer in the Solana community, and Andrew Kang of Mechanism Capita. Siong, one of the co-founders of jupiter is using a rootlet PPF as well.

1.3 Beyond Branding: The Product Comes First

Source: Suilend

While it’s undeniable that Suilend’s rapid growth was supported by the team’s prior experience building large-scale lending protocols on Solana, and by Rooter’s effective personal branding strategy, these factors primarily served as powerful tools for initial user acquisition. However, when it comes to building long-term success, nothing is more important than the product itself. Branding and track record may convince users to try an app—but if the product fails to deliver, users will churn just as quickly as they arrived. If retention is the most critical metric for evaluating a DeFi product’s success, then product quality is the foundational requirement for sustaining that retention.

In this respect, Suilend has consistently positioned itself at the forefront of innovation on Sui, often being the first to integrate many of the network’s most unique capabilities. For instance, it was an early adopter of NFT-based lending positions (a feature that has since become common across protocols), enabled seamless asset swaps and instant deposits within a single transaction using Sui’s Programmable Transaction Blocks (PTBs), and even launched its own LST (Liquid Staking Token) product—SpringSui—which is now deeply integrated into the Suilend deposit markets. The team has demonstrated remarkable execution velocity, shipping new features and upgrades on a near-weekly basis, all while keeping them tightly coupled to the core lending product to drive ongoing user engagement and improve retention.

Thanks to these efforts, Suilend—despite launching later than many of its DeFi peers on Sui—was able to establish itself as one of the ecosystem’s flagship protocols. This success can be attributed to a rare combination of rapid execution, agile branding, and consistently high product quality. But the team hasn’t stopped there. With the recent launch of STEAMM, a decentralized exchange, Suilend is now expanding its footprint to encompass a broader range of DeFi primitives.

What makes STEAMM particularly interesting is that it is tightly linked to Suilend at the design level. While the two protocols serve distinct functions—lending and trading—they are strategically integrated such that success in one directly benefits the other. This symbiotic structure follows a familiar pattern in how Rooter and the team have historically launched new products: not as isolated apps, but as interconnected components within a cohesive DeFi stack. It’s this system-level thinking that makes their approach stand out—not simply building multiple apps, but constructing a composable, vertically integrated DeFi ecosystem on Sui.

2. Suilend’s Bigger Picture: STEAMM

Source: STEAMM

At first glance, STEAMM may appear to be just another decentralized exchange (DEX), but a closer look at its architecture reveals several noteworthy design choices. Most notably, STEAMM introduces the first superfluid AMM on Sui—a concept we will explore in more detail later—and implements a novel market-making model called OMM (Oracle Market Maker). This model enables significantly more efficient pricing for large-volume trades compared to traditional CPMM-based systems. Together, these elements mark a clear departure from conventional DEX architecture and represent a compelling case study from a research standpoint.

Let’s now take a closer look at the core mechanisms that make STEAMM unique.

2.1 Enhancing Liquidity Availability via Interoperability: The Superfluid AMM Model

Source: STEAMM Doc

The crypto industry has long grappled with a fundamental question: how can we put idle liquidity to more capital-efficient use? One of the most prominent discussions on this topic was covered in my earlier research on liquidity availability, which explored how unused liquidity trapped within a single layer of infrastructure might be mobilized and redirected to applications that genuinely need it. As I noted in that article, solving liquidity availability at the infrastructure level is a notoriously difficult challenge.

However, the teams behind Suilend and STEAMM have taken meaningful steps toward addressing this issue by introducing a novel design: the Superfluid AMM. This mechanism, made possible by the seamless composability between their applications, provides a glimpse into how liquidity availability—once thought to be structurally constrained—can be unlocked through intentional protocol design.

In traditional AMM models, idle liquidity—funds not actively engaged in swaps—generates no additional yield and effectively becomes an opportunity cost. By contrast, STEAMM’s Superfluid AMM routes this idle liquidity into Suilend’s lending pools, allowing it to earn additional yield even while sitting passively in the AMM. This represents a meaningful step forward from the original AMM incentive model, where liquidity providers (LPs) were rewarded solely through trading fees. In the Superfluid AMM, LPs not only earn swap fees but also accrue interest generated by lending out idle assets, resulting in a significantly higher effective yield.

This added layer of incentives makes STEAMM’s liquidity pools inherently more attractive to LPs compared to conventional AMMs. As returns increase, more liquidity naturally flows into these Superfluid pools, reinforcing a positive feedback loop that can give STEAMM a strategic edge in the competitive DEX landscape—where liquidity is the most critical determinant of success.

Ultimately, the Superfluid AMM is a product that could only exist because Suilend and STEAMM were developed in tandem. It is a direct outcome of leveraging shared protocol infrastructure and a vertically integrated product strategy. Moreover, this design takes full advantage of Sui’s monolithic architecture, where smart contracts can interact with each other in a seamless, low-latency manner. The ability to compose lending and swapping functionalities atomically and trustlessly is what enables this kind of cross-functional capital efficiency—something that would be far more difficult on more fragmented or modular blockchain architectures. Without Suilend, STEAMM alone couldn’t have offered this; without Sui, the composability wouldn’t have been possible.

2.2 A Modular AMM Framework: How STEAMM Optimizes for Swap Intent

While STEAMM’s seamless integration with Suilend certainly provides a structural edge over traditional DEX models, this is far from its only differentiator. One of the key reasons STEAMM is increasingly regarded as a “next-generation DEX” lies in its modular Quoter architecture, which is engineered to optimize trading efficiency based on the type of asset and the purpose of the pool.

To understand this system, it’s important to first define what a Quoter is. In an AMM (Automated Market Maker) model, a Quoter is the component that estimates the expected exchange rate and slippage for a given swap. Since each liquidity pool is composed of different assets and uses a distinct pricing curve, using a one-size-fits-all Quoter across all pools would inevitably introduce inefficiencies. To address this, STEAMM utilizes three purpose-built Quoters, each optimized for a specific use case: CPMM, vCPMM, and OMM.

2.2.1 CPMM — For Standard Token Swaps

The Constant Product Market Maker (CPMM) is the most traditional AMM design, built on the familiar x * y = k formula. In CPMM-based pools, both assets in the trading pair must be present, and the model is most efficient when used for volatile token pairs, where price discovery is expected.

2.2.2 vCPMM — For Token Launches

he “v” in vCPMM stands for “virtual”, referring to a structure where a liquidity pool can be initialized with only one side of the asset pair by setting up a virtual reserve. While it still follows the traditional x * y = k constant product formula, trades can be executed based solely on one asset, making it a practical model for launching meme coins or new tokens.

One of the key advantages of vCPMM is that, unlike traditional CPMMs, it doesn’t require the quote asset to set the initial price. This allows token issuers to freely define the starting price without depositing both assets, which significantly mitigates the risk of sniping—a common issue during new token launches.

A good example of this is the launch of $ATTN, a token released by the InfoFi team using the vCPMM model. By doing so, they were able to launch the token more safely and securely compared to traditional launchpads like Pump.fun, with much lower exposure to sniping attacks.

2.2.3 OMM — For Large-Scale and Low-Slippage Trading

The Oracle Market Maker (OMM) represents a more advanced market-making model. Instead of relying on internal pool reserves for price discovery, OMM uses real-time price feeds from the Pyth oracle network, paired with a dynamic spread calculated based on order size and asset volatility. Pyth updates prices every 0.25 to 1 second. When a user submits a trade, the system evaluates the input size and market volatility, calculates the spread, and combines it with the oracle price to produce a final execution rate. A dynamic fee is applied—consisting of a base fee plus a portion of the spread—and the trade is executed in a single signature via Sui wallet.

The OMM structure offers multiple advantages over traditional AMM models. First, because the price is sourced from an oracle rather than determined solely by internal liquidity ratios, impermanent loss is significantly reduced. Second, slippage is minimized, especially for large trades, resulting in more stable execution.

To make OMM easier to understand, it can be compared to Uniswap V3’s concentrated liquidity model. In Uniswap V3, if the market price moves outside of a position’s specified range, the liquidity is no longer active, which means positions require constant monitoring and active management. In contrast, OMM automatically adjusts its quoting in real time based on the market price, allowing users to provide liquidity passively without needing to actively manage their positions.

Through this triple-Quoter design, STEAMM is able to provide optimized trade execution based on asset type and transaction context, delivering real efficiency gains for token issuers, traders, and liquidity providers alike. More than just another DEX, STEAMM positions itself as a flexible and modular DeFi platform capable of supporting a broad range of use cases through its composable design.

3. ONLY POSSIBLE ON SUI: What STEAMM Reveals About Sui’s Strengths

The story of STEAMM and its integration with Suilend offers a compelling answer to a fundamental question: Why did Mysten Labs decide to build an entirely new Layer 1 blockchain? Sam Blackshear, the CTO of Mysten Labs and one of Sui’s core contributors, once remarked that “the utility of a smart contract platform comes from its ability to allow stateful programs to interact with one another atomically.” In many ways, STEAMM’s superfluid AMM, where two independent protocols interact seamlessly at the smart contract level, is a concrete realization of that vision—and a testament to the unique capabilities of the Sui platform.

Indeed, the core functionalities that define STEAMM could not have been implemented without Sui’s distinctive architecture. Let’s take a closer look at the specific technical elements that made it all possible.

3.1 Atomicity via PTB (Programmable Transaction Block)

If you asked developers and users what sets Sui apart, many would point first to PTB, or Programmable Transaction Block. Simply put, PTB allows up to 1,024 sequential Move function calls to be executed in a single atomic transaction, all finalized with a single signature and a single gas payment. Either all steps succeed and are committed—or none are.

Other smart contract platforms, such as Solana, do support transaction batching. However, Solana enforces compute unit (CU) limits to protect network stability. These constraints often make it difficult to execute large, complex atomic transactions. Sui, on the other hand, employs a prepaid gas model that eliminates the need for CU limits while still securing the network. This architectural decision gives developers far more freedom to build multi-step, composable applications.

If one were to implement the same structure used by STEAMM on Solana, it would require at least three separate transactions due to Solana’s transaction size limits. However, since the logic must be executed atomically, the only viable way to achieve this on Solana is by using Jito bundles. The challenge is that Jito bundles are only included in a block when the current slot leader is running Jito, meaning there’s no guarantee that the bundle will actually be included in a timely manner. This introduces latency and unpredictability, which can degrade the overall user experience.

As a result, replicating this kind of architecture on Solana not only demands more complex transaction orchestration, but also involves external dependencies and timing constraints that make the process significantly more cumbersome and inefficient compared to building on Sui.

In STEAMM’s case, PTB is what makes its superfluid AMM truly viable. When a user executes a swap, STEAMM automatically detects the idle remainder of the swapped asset, routes it into Suilend’s lending pool, and completes both the trade and the deposit—all within a single signature and transaction.

3.2 Low-Latency Finality with Mysticeti Consensus

Even with atomic execution, fast finality is essential for capturing value from real-time composability. That’s where Sui’s consensus engine, Mysticeti, comes in. Mysticeti enables sub-second transaction finality, with latencies around 400ms, allowing protocols like STEAMM to act immediately on asset flows and redirect idle balances without delay.

This low-latency environment isn’t just a UX benefit for end users—it’s a core infrastructure enabler. In STEAMM’s case, it’s what makes the superfluid AMM logic not just theoretically possible but practically performant.

3.3 Object-Centric Model and Parallel Execution

Sui’s object-centric data model also plays a critical role. In this system, liquidity pools are treated as shared objects, while individual LP positions are owned objects. Because of this separation, thousands of concurrent swaps and deposits can be executed in parallel without transaction conflicts.

This stands in stark contrast to EVM-based systems, where state-sharing through global storage (like account balances or mapping slots) frequently causes execution bottlenecks due to lock contention. On Ethereum or similar platforms, it would be far more difficult—if not impossible—to implement STEAMM’s features without encountering significant throughput or consistency issues.

By leveraging Sui’s core architectural primitives—PTB for atomicity, Mysticeti for low-latency execution, and object-based parallelism—STEAMM and Suilend together showcase the full design intent behind Sui. Alongside protocols like Bluefin, they are not only flagship DeFi applications on the network, but also living demonstrations of what is only possible on Sui.

4. Why I’m Paying ATTention to Suilend

I’ve spent a long time researching and writing about the Sui ecosystem—but more than just a researcher, I’ve actively participated as a user over the past two years, trying out countless applications firsthand. And through that experience, I’ve come to a clear conclusion: Suilend and the team behind it are worth paying close and continuous attention to.

Suilend, in fact, represents only the starting point of a much broader vision. Since its launch, the team has experimented with and shipped a wide array of products: Rootlets, an NFT project with a strong community focus; SpringSui, the leading LST product on Sui; STEAMM, the new DEX with deep protocol integration; and most recently, the highly discussed information product, Attention. With each of these, the team has played an outsized role in driving energy and innovation across Sui’s DeFi ecosystem.

To some, this expansion into multiple verticals might seem unfocused or scattershot. But the data tells a different story. Suilend remains the most dominant lending protocol on Sui by TVL. Rootlets ranks as the third-largest NFT project by market cap. SpringSui holds the top spot in the LST category. And STEAMM, despite being a relatively late entrant to the DEX landscape, has already climbed to sixth place by trading volume on Sui. These are not isolated wins—they are the result of deliberately connected initiatives that have been architected with synergy in mind.

For example, Rootlets NFTs serve as a foundation for community engagement and can be linked to Suilend functionality. SpringSui assets are integrated directly into Suilend’s lending markets. And idle assets generated through STEAMM’s superfluid AMM are routed back into Suilend for additional yield generation. These systems are not operating in silos—they are mutually reinforcing components of a vertically integrated strategy. In this sense, the team behind Suilend may be among the best in the industry at understanding and operationalizing the structural advantages of a monolithic blockchain like Sui.

Over time, this strategic integration creates a deep and defensible moat—one that no competitor can easily fork or replicate. And that is precisely why I intend to continue watching what this team builds next. The more synergistic their stack becomes, the more irreplaceable they will be in the Sui ecosystem.