How HumidiFi Became Solana's Largest Prop AMM

The rise of proprietary automated market makers (prop AMMs) represents one of the most significant shifts in decentralized trading infrastructure over the past year. At the center of this transformation on Solana sits HumidiFi, a protocol that went from first trade to largest DEX by volume in just a matter of months. In a revealing conversation on Lightspeed, Kevin, a former Jump Trading veteran who now leads HumidiFi, pulled back the curtain on how these sophisticated trading systems work, why they've captured the majority of trading activity on Solana, and what this means for the future of on-chain price discovery.

Top Projects

Lulo

Score: 7,586.13

Bonk

Score: 4,806.12

Sphere

Score: 2,544.42

Moonwalk

Score: 2,026.11

The Genesis of HumidiFi

HumidiFi emerged from a combination of traditional finance expertise and cutting-edge blockchain engineering. Kevin spent approximately nine years at Jump Trading, one of the most well-known high-frequency trading firms in traditional markets, before transitioning into cryptocurrency about five years ago. His journey through crypto included time at Paradigm and a trading firm called Symbolic Capital Partners before the opportunity to build HumidiFi presented itself.

The origin story of HumidiFi is surprisingly organic. Kevin was trading on Solana, running automated strategies, but finding limited success with existing infrastructure. Around six or seven months ago, he connected with the Temporal team and discovered they had complementary skill sets. After meeting in person and spending a weekend together, the first minimum viable product of HumidiFi was born. The rapid progression from that initial prototype to becoming Solana's dominant trading venue speaks to both the team's execution capabilities and the hungry demand for better trading infrastructure on the network.

"We weren't really expecting to hit the sort of volume metrics that we have, but that's not really been the focus," Kevin explained, highlighting that the team has prioritized building superior technology over chasing vanity metrics. The results, however, speak for themselves: HumidiFi completed its first trade in June, became the number one prop AMM by August, and achieved the status of largest DEX on all of Solana by October.

Understanding Proprietary AMMs

Traditional AMMs like Raydium, Meteora, and Orca are designed to programmatically allow users to deposit liquidity and generate trading pool fees. The model is relatively passive—liquidity providers set their parameters and let the smart contract handle the rest. Prop AMMs operate on fundamentally different principles that enable them to offer dramatically better prices to traders.

Rather than accepting customer deposits, prop AMMs like HumidiFi use their own proprietary capital to provide liquidity. More importantly, they employ sophisticated trading algorithms that update quotes in real-time, responding to market conditions with a level of responsiveness that passive AMMs simply cannot match. While traditional AMMs might update their pricing parameters occasionally or rely on static bonding curves, HumidiFi sends multiple price updates per block, constantly streaming fresh quotes to ensure they're offering competitive prices.

"Every single block, actually multiple times a block, we're streaming in different updates to update the prices that we're willing to trade at," Kevin explained. This constant updating mechanism is what differentiates prop AMMs from their passive counterparts and enables them to capture such a significant portion of trading volume.

The terminology around these protocols has evolved as understanding has grown. Initially, they were called "dark AMMs" on crypto Twitter because nobody understood how they worked. The opacity led to speculation and confusion. Now, with greater transparency and education efforts, the community has developed a better understanding of the sophisticated technology underlying these protocols.

The First Generation of Prop AMMs

HumidiFi didn't invent the concept of prop AMMs on Solana. Kevin was careful to acknowledge the pioneering work of the first cohort: Solphi, Obrick, and ZeroFi. These protocols created the next-generation platform that demonstrated the viability of the approach. Even before them, Affinity attempted to pioneer something using similar technologies, specifically employing an off-chain oracle sender to update prices in a more real-time manner, using Pyth as an oracle.

The key innovation that separated the first generation of prop AMMs from Affinity's approach was the use of proprietary pricing models rather than relying on external oracles. Each prop AMM developed its own view on what prices should be, creating a competitive landscape where better price models could capture more volume. This competition has driven continuous innovation in pricing accuracy, compute efficiency, and transaction landing capabilities.

The rapid market share capture wasn't a surprise to Kevin. "If that's kind of the beauty of crypto and especially on Solana, where ultimately, it's only performance that matters," he noted. "And if you're able to offer a better product within a couple months, you could actually acquire a lot of market share and become like one of the top narratives."

Capital Efficiency as a Competitive Advantage

One of the most striking aspects of HumidiFi's operation is the extreme capital efficiency it achieves. Looking at the numbers, HumidiFi maintains a total inventory or TVL of only around 8 million dollars, yet regularly processes 500 million to over a billion dollars in daily volume. This ratio would be impossible for traditional AMMs and represents a fundamental rethinking of how liquidity provision can work.

The secret lies in concentrated, just-in-time liquidity. With concentrated liquidity AMMs like Uniswap V3, traders began placing liquidity at the top ticks near the current price. Prop AMMs take this concept to its logical extreme: all liquidity is placed just for the next trade. There's no capital sitting idle waiting for unlikely price scenarios.

"If people want to be able to sell all their Solana down to like a price of $1, we won't be there. We have no liquidity waiting to buy Solana at $1," Kevin explained. "However, if you want to sell Solana right now for as much as if you want to sell one SOL, then we're going to be able to provide you the absolute best price."

This approach means HumidiFi can offer superior pricing for typical retail trades while maintaining profitability. The trade-off is that they're not providing the deep liquidity reserves that might be needed in extreme market conditions—but for the vast majority of trading activity, this capital efficiency translates directly into tighter spreads for users.

The Pricing Model and Profitability

At the core of HumidiFi's operation is a predictive price model—the kind of sophisticated system that firms like Jump or Tower would use in traditional finance. This model aggregates data from multiple sources: Binance, OKX, Coinbase, and other major centralized exchanges, looking at both spot and perpetual markets. It also incorporates data from decentralized exchange venues to capture information that might not be available through centralized sources.

The goal is to determine the "fair value" of an asset at any given moment and stream that information onto the Solana blockchain as efficiently as possible. Once the oracle update lands on-chain, HumidiFi offers prices to users: willing to buy below the oracle price (the bid) and sell above it (the offer) for any size trade on any asset they quote.

"The difference between that Oracle price update and the price that we transact at, that's what we consider our profit," Kevin revealed. In theory, every trade where HumidiFi successfully lands an oracle update is profitable. However, the reality is more nuanced—there are sophisticated competitors and arbitrageurs who have their own price models and execution capabilities.

When analyzing profitability, HumidiFi looks at metrics like markouts—how prices move after trades are executed—as well as the total accumulation of funds in their pools. The team has extensive analytics on different market participants, profiling bots, centralized exchange traders, and retail traders to understand the flow they're handling.

The Philosophy of Trading Against Everyone

HumidiFi has taken what Kevin described as a "bold stance" on handling adversarial flow. Unlike some competitors who have chosen to blacklist or refuse to trade against informed traders who consistently profit at the market maker's expense, HumidiFi has committed to improving their systems to remain profitable regardless of who's asking for a quote.

"We're competitive and we don't like being beat by anybody," Kevin explained. "So when we see that somebody's able to consistently profit off of us, we try to figure out why. And so far, we've been able to kind of combat every adversary."

This approach serves multiple purposes. It drives internal improvement, forcing the team to continuously enhance their pricing models and execution capabilities. It also maximizes the volume HumidiFi can capture. The arbitrage trades between HumidiFi and other venues like Orca or Solphi are still profitable from HumidiFi's perspective, and they help the broader ecosystem maintain price consistency across venues.

The distinction between toxic flow (sophisticated traders extracting value) and retail users seeking the best price isn't black and white. By maintaining the capability to trade profitably against anyone, HumidiFi naturally provides better prices to retail users while still capturing profitable arbitrage opportunities. The team hasn't ruled out eventually implementing some form of flow discrimination, but for now, their competitive nature drives them to be "the alpha on the chain."

Moving Price Discovery On-Chain

Perhaps the most ambitious long-term goal for prop AMMs is facilitating genuine price discovery on the blockchain rather than merely copying prices from centralized exchanges. Dan Smith of Blockworks Research has been tracking data showing that significant portions of trading volume for new tokens are flowing through prop AMM venues—sometimes 60-80% of total volume for assets like the MON token.

Kevin confirmed this is exactly the direction HumidiFi wants to push. "Just taking Binance prices, just copying Binance mid onto the blockchain is not actually providing a service to anybody," he noted. The goal is to help create liquidity for assets where people don't necessarily want to trade on centralized exchanges, doing so more efficiently than passive AMMs can.

The team has an internal joke about "abstinence trading"—trading with no centralized exchange involvement. The concept involves creating price models that only look at decentralized trading activity. Traditional HFT firms have solved similar problems; there's always some form of price discovery happening, and HumidiFi aims to become that source for Solana across as many assets as possible.

Orca was the first token HumidiFi tried to quote without centralized exchange listings. It's particularly challenging because the token moves significantly and has limited trading venues. But this is exactly the kind of experimentation needed to push decentralized trading forward. If Solana wants to be the home of internet capital markets, it needs to support the assets people want to trade—and support them from minute one of trading.

The Monad Launch: A Case Study in Day-One Liquidity

The recent launch of MON (Monad's token) provided a compelling demonstration of what prop AMMs can achieve. Wormhole had created Project Sunrise and reached out to HumidiFi on a Sunday night, just hours before the launch. Despite the short notice, the team had their systems running within an hour of when MON started trading on Monad itself and on Coinbase.

The results exceeded expectations. By the end of the launch day, HumidiFi was one of the top venues for MON trading in the entire crypto space. Within a couple of hours, they were offering better prices than any venue, including centralized exchanges and DEXs on Monad's own chain.

"We didn't really know what to expect in terms of the performance," Kevin admitted. "But yeah, by the end of the day, I think we were one of the top venues in the entire crypto space." This success demonstrated that Solana, powered by sophisticated trading infrastructure, can compete with and even outperform native chain venues for new token launches.

The MON launch exemplifies the vision of Solana as a day-one liquidity venue. If traders participate in a token sale on one platform, receive their assets, and then realize HumidiFi on Solana offers the best trading experience, they're likely to bring those assets to Solana rather than trading elsewhere. Winning the day-one competition creates sticky user behavior.

Technical Optimization: The CU Race

One of the most important metrics for prop AMM performance is compute unit (CU) usage per transaction. Data from Blockworks' new HumidiFi dashboard shows the protocol has decreased CU usage from around 300 to under 50 over the past six months—a dramatic improvement that directly impacts competitiveness.

In traditional markets, different trading firms have varying execution speeds based on their technology infrastructure. Some use FPGAs and ASICs to achieve microsecond advantages. On Solana, CU efficiency serves a similar purpose. When the network's sequencer is deciding whether to include one oracle update before another, lower CU usage can be a determining factor.

"For us being able to reduce from 300 to 47, this means that when the sequencer is deciding whether to put our Oracle update ahead of some other prop AMM's Oracle update, this will be one of the determining factors," Kevin explained.

The team's partners at Temporal specialize in this kind of optimization. The result is that HumidiFi can get oracle updates and prices on-chain faster than competitors, enabling them to quote tighter spreads with confidence. If they couldn't achieve this speed, they might face increased risk from adverse selection by other AMMs.

Handling Transaction Reverts

Not every oracle update succeeds, and understanding why helps illuminate how prop AMMs operate. HumidiFi tries to land every oracle update using multiple methods—some combination of TPU (Transaction Processing Unit) connections and Jito, the major MEV infrastructure on Solana.

Sometimes transactions revert simply due to network conditions. More problematic is when a price moves one direction, triggering an oracle update, but then reverses before that update lands. If the first update lands but the second doesn't, HumidiFi might be offering to buy at a price that no longer reflects fair value, creating a losing position.

For user transactions, reverts typically occur when slippage tolerance is set too low. If a user sees a quote to trade at $130 but SOL rises to $140 before their transaction executes, and their slippage tolerance doesn't accommodate that movement, the transaction fails. This is unrelated to HumidiFi's oracle updates—it's simply how trading works in volatile markets.

All of HumidiFi's technology optimizations, including the CU improvements and transaction landing capabilities through Temporal's Enzomi product, serve to allow safer and more aggressive quoting. By minimizing unprofitable trades, the protocol can pass those savings on to users through tighter spreads.

The October 10th Incident: A Skill Issue

The market volatility around October 10th created a moment of vulnerability that sparked considerable debate on crypto Twitter. Some prop AMMs, including HumidiFi, stopped quoting prices during the extreme volatility, leading to theories about intentional manipulation or fundamental flaws in the prop AMM model.

Kevin provided the actual explanation: their oracle sender and price prediction model crashed. The system simply couldn't handle the unprecedented volatility. Many other HFT strategies, including centralized exchange operations, faced similar failures. Making matters worse, Kevin was at lunch interviewing a candidate when the crisis hit, requiring him to race back to manually restart the system.

"I consider it like a skill issue. Like in our case, our system wasn't ready," Kevin acknowledged. The crash cost HumidiFi more than $500,000 in missed profit—exactly the kind of high-volatility conditions where their competitive advantages should shine. Competitors who maintained uptime did exceptionally well during that period.

The incident highlighted an important principle: while prop AMMs naturally widen spreads during high volatility (which is rational behavior that shouldn't be criticized), they shouldn't completely stop quoting. That's a system failure, not a design choice. HumidiFi has since improved their systems to better handle extreme conditions, though Kevin noted no system can guarantee 100% uptime forever.

The Case for Ecosystem Redundancy

Despite HumidiFi's dominance, Kevin strongly advocated for maintaining multiple liquidity sources in the ecosystem. No single market maker should be the only venue for trading—that would defeat the purpose of decentralization and create systemic risk.

"I don't think any sort of market microstructure should rely on one party," Kevin stated. "I think having other prop AMMs available, other AMMs in general available, and other centralized order books that are evolving is a very healthy thing for the ecosystem."

Competition drives improvement. When HumidiFi sees a competitor consistently profiting against them or performing better in certain conditions, it motivates internal improvement. The October 10th incident demonstrated that even market leaders can have failures, making redundancy essential for network resilience.

This philosophy extends to aggregators as well. The competitive landscape among Jupiter, DFlow, Titan, and OKX aggregators pushes each to improve, ultimately benefiting users. HumidiFi aims to support all aggregators equally and work with whichever ones ship new features fastest.

Solana vs. Other Chains: Why the Home Matters

The unique characteristics of Solana's blockchain architecture make it particularly well-suited for prop AMM operations. Ethereum L1's 12-second block times create very different dynamics—pricing accuracy matters more than latency since transactions only occur every 12 seconds. The market microstructure is better documented, and the MEV supply chain is well understood.

Solana's effective 15-millisecond sub-slots within 400-millisecond slots create a much more high-frequency environment. Being able to consume data and respond quickly, then land transactions effectively, remains "more of an art than a science." The rotating leader schedule means good connectivity is needed everywhere, unlike Ethereum where validator locations are more predictable.

Kevin revealed an industry secret: sophisticated HFT firms have been using dark fiber—dedicated private fiber optic connections—to move data around Solana more quickly. Double Zero's public discussion of this infrastructure "blew the spot for everybody," making it common knowledge that better networking provides meaningful advantages.

Base represents the closest competitor to Solana's microstructure among Ethereum L2s. Kevin noted that Tasera, one of HumidiFi's Solana competitors, appears to have launched a prop AMM on Base, proving the concept can work elsewhere. However, Base's single sequencer creates different dynamics—the target location for low-latency connections is well-defined, unlike Solana's distributed validator set.

For HumidiFi, Solana will always be home. The chain's teams are responsive to feedback about trading infrastructure needs, and there's alignment between what Solana is trying to achieve (becoming the home of trading) and what prop AMMs need to operate effectively.

The Aggregator Relationship

A significant portion of HumidiFi's volume flows through aggregators like Jupiter, DFlow, Titan, and OKX. This relationship has evolved over time and continues to develop. Initially, the dominance of Jupiter meant that connecting to their flow was sufficient to justify the work of building a prop AMM. Now the aggregator landscape is much more competitive.

HumidiFi's model involves maintaining an off-chain oracle, an on-chain smart contract, but no frontend interface. When users trade through an aggregator, the router queries HumidiFi (and other liquidity sources) for the best price for a given trade size. The aggregator then composes the optimal swap path across all available venues.

"Our customers are all the users on Solana," Kevin explained. "But if one aggregator is shipping new features better and faster than the others, then they're going to start winning. And we're definitely going to be happy to support that."

The relationship involves mutual adaptation. Initially, HumidiFi had to conform to Jupiter's requirements. As the ecosystem matures, Kevin expects more bidirectional influence—HumidiFi requesting aggregators support new features while aggregators push for new capabilities from prop AMMs.

Recent innovations demonstrate this dynamic. DFlow announced "jet routing" that checks prices in real-time rather than using stale quotes. Jupiter quickly implemented something similar. This competitive innovation benefits everyone in the ecosystem.

The WET Token Launch

HumidiFi is launching its WET token through Jupiter's new Decentralized Token Formation (DTF) platform—the first project to do so. The choice aligned with HumidiFi's values around transparency. Despite the team's secretive nature as proprietary traders, their personal and professional reputations matter deeply, making transparent token dissemination important.

The partnership with Jupiter made practical sense given their long relationship as aggregator and liquidity provider. But Kevin emphasized the DTF platform's merits stand independent of that relationship—the transparency about how tokens are distributed, where wallets are located, and what will happen aligns with HumidiFi's principles.

Being the first project on a new platform carries risk but also opportunity. HumidiFi wanted to "kick things off" for DTF, potentially establishing the platform as a legitimate launch mechanism for serious projects.

Looking Forward: Solana in 2025 and Beyond

When asked about exciting Solana developments, Kevin focused on the goal of becoming the primary day-one liquidity venue for new assets. Multiple things must come together: the ability to bring assets on-chain (like Wormhole's Project Sunrise), venues capable of trading without centralized exchange dependencies, and users who want to trade.

The chicken-and-egg problem is real but surmountable. Success requires playing for keeps—half-measures won't establish Solana as the home of trading. If traders participate in token sales elsewhere but discover Solana offers better trading, fragmentation across chains creates friction that hurts everyone.

Kevin mentioned MCP (Multi-Chain Protocol) as potentially six months to a year out, with various companies working on solutions. Rather than expressing strong opinions on which approaches will win, he appreciates the discourse and competition driving everyone to push harder.

The most important thing is Solana's continued emphasis on doing a few things well—specifically trading—rather than trying to do everything. This focus creates space for specialized protocols like HumidiFi to push the boundaries of what's possible.

The Broader Vision

HumidiFi's ambitions extend beyond capturing trading volume. The team sees themselves as "the technical trading-focused service provider that allows a lot of this stuff to happen"—infrastructure that enables the broader ecosystem rather than an all-in-one application.

Supporting Solana's goal of becoming internet capital markets requires supporting the assets people want, from launch day onward. Traditional markets have market makers; decentralized markets need prop AMMs and similar infrastructure to match that capability. The goal is providing deep liquidity and tight spreads whether someone is trading $10 or $10 million worth of tokens.

The competition isn't just against other Solana venues—it's against the entire crypto ecosystem. Can Solana provide better prices than Coinbase for new token launches? Can on-chain venues discover prices without relying on centralized exchange feeds? HumidiFi believes the answer to both questions can be yes, and they're building the technology to prove it.

For Solana observers, prop AMMs represent a maturation of DeFi infrastructure. The move from passive liquidity provision to sophisticated market making mirrors how traditional markets evolved. The difference is that on Solana, this evolution is happening in public, with transparent volume data and competitive dynamics visible to anyone watching.

As Kevin noted, ultimately only performance matters. If you build something people want, you can capture market share quickly regardless of how established competitors might be. HumidiFi's rise from weekend prototype to dominant trading venue demonstrates this principle. The question now is how far this competitive pressure can push on-chain trading—and whether Solana's bet on becoming the home of trading will pay off.

Facts + Figures

• HumidiFi completed its first trade in June 2024, became the #1 prop AMM by August, and the #1 DEX on Solana by October

• Kevin, HumidiFi's leader, spent approximately 9 years at Jump Trading before entering crypto about 5 years ago

• HumidiFi maintains around $8 million in TVL while processing $500 million to over $1 billion in daily volume

• Compute unit usage per transaction decreased from ~300 CUs to under 50 CUs over six months

• 60-80% of MON token trading volume occurred through prop AMM venues

• HumidiFi was offering better MON prices than any venue (including Coinbase) within hours of launch

• The October 10th volatility incident cost HumidiFi more than $500,000 in missed profit due to system crashes

• HumidiFi sends oracle updates multiple times per block, compared to passive AMMs' static parameters

• The protocol uses data from Binance, OKX, Coinbase, and other major exchanges for price modeling

• Solana has effective 15-millisecond sub-slots within 400-millisecond slots, enabling high-frequency trading

• Dark fiber networking provides meaningful advantages for sophisticated trading operations on Solana

• HumidiFi is launching its WET token through Jupiter's DTF platform—the first project to use the platform

• Tasera, a HumidiFi competitor, appears to have launched a prop AMM on Base

• HumidiFi quotes approximately 15 tokens beyond the primary SOL-USDC pair

• The team consists of Kevin, one other close collaborator, and engineers from the Temporal team

Questions Answered

What is a prop AMM and how does it differ from traditional AMMs?

A prop AMM (proprietary automated market maker) uses its own capital rather than customer deposits and employs sophisticated algorithms that update prices in real-time, multiple times per block. Traditional AMMs like Raydium, Meteora, and Orca allow users to deposit liquidity with set parameters that remain relatively static. Prop AMMs stream constant price updates based on predictive models that incorporate data from centralized exchanges, perpetual markets, and other venues. This enables them to offer dramatically tighter spreads and use capital far more efficiently than passive liquidity pools.

How can HumidiFi process billions in volume with only millions in TVL?

The extreme capital efficiency comes from placing all liquidity just-in-time for each trade rather than having capital sit idle at various price levels. Traditional AMMs spread liquidity across wide price ranges, meaning most capital is unused at any given time. HumidiFi's approach means no liquidity waits at unrealistic prices—the protocol won't be there to buy SOL at $1—but it provides the absolute best prices for trades at current market levels. This concentrated, just-in-time liquidity model means every dollar of capital is actively deployed for the next trade.

How does HumidiFi make money on trades?

HumidiFi runs an off-chain predictive price model that determines fair value for assets by aggregating data from Binance, OKX, Coinbase, and other major venues. This fair value is sent as an oracle update to the Solana blockchain. The protocol then offers to buy below that price (the bid) and sell above it (the offer). The spread between the oracle price and the actual transaction price represents profit. Every trade where an oracle update lands successfully should theoretically be profitable, though sophisticated competitors can sometimes beat HumidiFi with better price models or faster execution.

Why do prop AMMs widen spreads or stop quoting during high volatility?

During extreme volatility, the risk of offering stale prices increases dramatically. If an oracle update hasn't landed recently, offering tight spreads could result in significant losses from adverse selection. Widening spreads during volatility is rational behavior that reflects uncertainty—it shouldn't be criticized. However, completely stopping quotes (as happened to HumidiFi on October 10th) represents a system failure, not a design choice. HumidiFi's price model crashed under unprecedented conditions, costing them over $500,000 in missed profit during conditions where they should have thrived.

Why doesn't HumidiFi blacklist sophisticated traders who profit against them?

HumidiFi takes a competitive stance: rather than blacklisting adversarial flow, they improve their systems to remain profitable regardless of who's trading. This approach serves multiple purposes—it drives internal improvement, maximizes volume capture, and helps maintain price consistency across venues. When arbitrageurs trade between HumidiFi and other venues, those trades are still profitable for HumidiFi while helping align prices across the ecosystem. The philosophy reflects the team's competitive nature: "we don't like being beat by anybody."

Can price discovery actually move on-chain away from centralized exchanges?

Yes, and this is HumidiFi's ambitious goal. Simply copying Binance mid prices onto the blockchain doesn't provide real value. The team is working on "abstinence trading"—quoting assets using only decentralized trading data without centralized exchange inputs. They've already experimented with Orca, which has no centralized exchange listings. Traditional HFT firms have solved similar problems; there's always price discovery happening somewhere. HumidiFi aims to become that price discovery source for Solana across as many assets as possible.

Why is Solana particularly well-suited for prop AMMs compared to other chains?

Solana's 15-millisecond sub-slots within 400-millisecond blocks create a high-frequency trading environment where speed matters enormously. Ethereum's 12-second block times make pricing accuracy more important than latency. Solana's architecture requires good connectivity everywhere due to rotating leader schedules, rewarding sophisticated networking infrastructure like dark fiber. The chain's development teams are also responsive to trading infrastructure needs, creating alignment between Solana's goal of becoming the home of trading and what prop AMMs need to operate effectively.

How important are aggregators to HumidiFi's success?

Aggregators like Jupiter, DFlow, and Titan are crucial partners—HumidiFi has no frontend and relies on aggregators to route user trades to their liquidity. Initially, Jupiter's dominance meant that connecting to their flow justified building a prop AMM. Now the competitive aggregator landscape pushes everyone to improve. HumidiFi adapts to aggregator requirements while also expecting to influence aggregator development as the ecosystem matures. The relationship is symbiotic: aggregators need best-price liquidity, and prop AMMs need user access that aggregators provide.