From "On-Chain Applications" to "Financial Infrastructure": The Generational Evolution and Transformation of Perp DEX

marsbit2026-01-15 tarihinde yayınlandı2026-01-15 tarihinde güncellendi

Özet

Title: From "On-Chain Applications" to "Financial Infrastructure": The Generational Evolution of Perp DEX The 2025 period was a "great filtering era" for the derivatives sector. Surviving Perp DEX platforms have moved beyond being mere "low-cost versions" of CEXs by solving the core cost in finance: trust. With the adoption of full-chain abstraction, users in 2026 can perform seamless cross-chain transactions while retaining asset sovereignty, as funds are secured in smart contracts rather than held by intermediaries. On-chain derivatives now consistently account for over 25% of total trading volume, marking a fundamental shift in user behavior. Over 90% of Perp DEXs failed due to product homogeneity, reliance on subsidized "rented liquidity," and soaring customer acquisition costs. Merely forking existing code or offering token incentives proved unsustainable. Four successful models have emerged: 1. **Hyperliquid**: Achieved near-CEX performance by building its own L1 blockchain optimized for low-latency order books, attracting quantitative capital. 2. **Aster**: Leveraged the Binance ecosystem to offer enhanced capital efficiency, allowing users to earn yield on collateral (e.g., staking rewards) while trading. 3. **Lighter**: Built an app-specific ZK-Rollup to provide a verifiable, mathematically-proven trading infrastructure with anti-MEV properties, appealing to institutions. 4. **Decibel**: Unified high performance and full-chain composability on Aptos, achieving...

Author: Max.s

The year 2025 is often regarded as the "Great Filter Era" for the derivatives sector. As we stand before the current power landscape of Perp DEX, it is clear: the vast majority of once-prominent forked projects have faded into obscurity, while the surviving ones are reshaping the financial order with a全新的的姿态.

The survival of Perp DEX does not stem from being faster or cheaper than centralized exchanges (CEX), but from addressing the core cost in the financial system — trust. After several crises involving opaque清算 mechanisms in second-tier exchanges (the Double-Ten crash left all market makers in agony!), the market has reached a consensus: transparency is not an option but the foundational logic of infrastructure.

Early Perp DEXs were often seen as "low-end versions" of CEXs. However, with the普及 of full-chain abstraction (Chain Abstraction) technology, users in 2026 can achieve seamless cross-chain experiences. User assets no longer need to be custodied by intermediaries but are locked in smart contracts. This return of "asset sovereignty" is the底气 for Perp DEX to claim a share of the market from CEXs.

Currently, on-chain derivatives trading volume has stably captured over 25% of the total market share. This is not merely numerical growth but a migration in user behavior patterns. When清算 logic, funding rates, and order matching are recorded on an immutable ledger, Perp DEX has evolved from experimental DApps to essential infrastructure in the crypto market.

The demise of most Perp DEXs: mediocrity is the original sin. Behind the prosperity lies an极其残酷 elimination race. Over the past two years, more than 90% of Perp DEXs have fallen into silence. The "causes of death" for these failures are highly consistent: product homogeneity, reliance on subsidies for liquidity, and lack of technical depth.

During the "Points" frenzy, numerous projects attracted users with inflated liquidity mining rewards. However, once points were redeemed and airdrops distributed, these platforms fell into a death spiral of "liquidity draining to zero." This model reliant on rented liquidity will gradually disappear from professional markets starting in 2026.

Note: Rented liquidity refers to the model where DEXs incentivize users to provide capital to support trading depth by distributing points or token subsidies. Simply put, the protocol does not truly "own" this liquidity but temporarily rents users' funds by paying high "rent" (token or point rewards).

Another reason for the downfall of DEXs is the soaring customer acquisition cost. In the absence of a unique niche, merely tweaking a UI or forking GMX's code is no longer sufficient to survive in this highly competitive market. Projects without core advantages in order matching engines or strong ecosystem backing are essentially expensive liquidity mining pools, not real exchanges.

Four existing DEX models值得后来者们借鉴学习:

Hyperliquid Model: Vertical Integration and Technological Hegemony

On the list of surviving DEXs, Hyperliquid is an unavoidable monument. It proves that if general-purpose公链 cannot support high-frequency trading, the best solution is to build your own chain.

Hyperliquid attracts substantial quantitative capital because it addresses order book latency through L1底层 optimization. It no longer寄生 on Arbitrum or other Layer 2 solutions but has built a consensus mechanism specifically designed for derivatives. This "vertical integration" grants it matching performance close to CEXs while maintaining on-chain transparency.

More importantly, Hyperliquid has successfully built a "quant-friendly" ecosystem. When third-party market makers发现这里的 API latency is extremely low and slippage is controllable, endogenous liquidity begins to grow organically. This "performance barrier" constructed through hard technological strength makes it从容 in facing competition from generic DEXs lacking distinctive features.

Aster Model: Ecosystem Premium and Asset Management Layer

If Hyperliquid relies on hardcore technology, then Aster and its backing from the Binance ecosystem represent another survival logic:极致 resource efficiency and asset appreciation — essentially, having a strong "backer."

Aster is not just a trading venue; it更像是一个 "leveraged layer for yield-bearing assets." Through deep integration with the Binance ecosystem, it introduces collateral like asBNB or USDF, allowing users to earn Staking or Re-staking yields while holding positions. This optimization of capital efficiency is难以企及 for standalone DEXs.

For large capital users, funding cost is a core consideration. When users open a position on Aster, their margin continues to generate annualized yields. This logic of "liquidity assetization" turns Aster into a highly sticky financial gateway, not merely a speculative tool.

Lighter Model: ZK-Powered Verifiable Financial Infrastructure

The Lighter model represents the pinnacle of "financial infrastructure-ization." It does not seek to become a traffic入口 but provides institutions with a交易底层 offering mathematical certainty through its self-built application-specific ZK-Rollup (App-specific ZK-rollup).

Lighter's uniqueness lies in solving the "mathematical honesty" problem. It encodes order matching and清算 logic into "ZK circuits." This means that the matching and清算 of every trade no longer rely on the "reputation" of nodes but on verifiable mathematical proofs. This is extremely attractive to institutional investors who despise "black-box清算."

Furthermore, Lighter's ZK-Orderbook design inherently possesses anti-MEV properties, protecting the strategy privacy of high-frequency traders. This combination of "verifiability + privacy protection + extremely low latency" makes it a standard interface linking real-world assets (RWA) with on-chain derivatives, building a high compliance and technological moat.

Decibel Model: Unification of Extreme Performance and Full-Chain Composability

In the 2026 market, Decibel represents the third evolutionary model for the new generation of Perp DEXs: the乐高 combination of a "high-performance engine" and "composability." As a full-chain trading engine rising within the Aptos ecosystem, Decibel彻底终结了 the fate that "speed and decentralization cannot be兼得".

Decibel's core competitiveness lies in its deeply optimized Trading VM. Leveraging Aptos's Block-STM parallel execution architecture, it is advancing towards sub-20 millisecond (sub-20 ms) block times and the ability to process over 1 million orders per second. This makes on-chain order matching no longer an "illusion" but a reality capable of competing with top-tier CEXs.

Unlike traditional isolated DEXs, Decibel provides a highly programmable financial platform. It unifies spot trading (Spot), perpetual contracts (Perps), margin trading (Margin), and vaults (Vaults). This "full-stack" design means users can use a single cross-chain margin account to collateralize various assets like APT, USDC, BTC, ETH simultaneously, greatly释放了 capital efficiency.

Decibel's "X-Chain Accounts" technology further breaks down cross-chain barriers. Users can directly fund their accounts using wallets from Ethereum or Solana (like MetaMask or Phantom) without configuring complex cross-chain bridges. This "seamless access" capability, combined with 100% on-chain matching logic, positions Decibel to potentially become the new favorite for on-chain high-frequency traders and institutions in 2026.

New Directions Post-2026: Intent, AI, and Dynamic Pricing

From an industry practitioner's perspective, the future evolution of Perp DEX will focus on the following three dimensions:

Intent-Centric Trading Experience. Future users will no longer need to manually adjust funding rates or slippage but will express an intent. The system will find the optimal execution path across chains through solvers. This model will significantly lower the barrier for retail users to access complex derivatives.

The Explosion of AI Agents. With the maturation of on-chain automation tools, DEXs will built-in AI strategy engines. A significant portion of future open interest will be driven by AI. This means DEXs need to provide more powerful computing capabilities and lower data latency to accommodate the high-frequency博弈 of robots.

Evolution of Pricing Models. Current AMMs or simple order books remain fragile during extreme market conditions. We are seeing more projects introduce complex dynamic risk engines that automatically adjust system parameters in real-time through more scientific formulas:

实时调节系统参数。This automated adjustment based on real-time volatility and position deviation will make the system robustness of Perp DEX truly surpass that of traditional centralized institutions.

The second half for Perp DEX is a survival race about "efficiency." Projects that attempted to sustain themselves through mediocre subsidies have long turned to dust. The future winners will either possess impeccable technological foundations like Hyperliquid, have irreplicable ecosystem resources like Aster, or have found the perfect balance between performance limits and full-chain composability like Decibel.

In this field, there is only one reason to survive: Do you provide an execution efficiency that capital and strategies cannot refuse?

İlgili Sorular

QWhat is the core value proposition of Perp DEXs that allows them to compete with CEXs, according to the article?

AThe core value proposition is not being faster or cheaper than CEXs, but rather solving the most fundamental cost in the financial system: trust. Perp DEXs provide transparency, with清算 logic, funding rates, and order matching recorded on an immutable ledger, which has become a consensus requirement after crises involving opaque清算 on second-tier exchanges.

QWhat is 'Liquidity Renting' and why did it lead to the failure of many Perp DEX projects?

ALiquidity Renting refers to a model where a DEX induces users to provide capital to support trading depth by distributing points or token subsidies. The protocol doesn't truly 'own' this liquidity but temporarily rents users' funds by paying a high 'rent' (token or point rewards). This model failed because once the points were redeemed and airdrops distributed, these platforms fell into a 'liquidity to zero' death spiral, as the liquidity was not sustainable without continuous subsidies.

QHow does the Hyperliquid model achieve its competitive advantage in the Perp DEX space?

AHyperliquid achieves its advantage through vertical integration and technical hegemony. It built its own L1 blockchain specifically optimized for derivatives trading, solving the order book latency issues of general-purpose blockchains. This gives it CEX-like matching performance while maintaining on-chain transparency. It also created a 'quant-friendly' ecosystem with low API latency and controllable slippage, fostering endogenous liquidity growth and building a strong performance barrier.

QWhat unique problem does the Lighter model solve with its ZK technology, and for which user base is it particularly attractive?

AThe Lighter model solves the problem of 'mathematical-level honesty' by building an application-specific ZK-Rollup. It encodes order matching and清算 logic into ZK circuits, meaning every trade's execution and清算 relies on verifiable mathematical proofs rather than the 'reputation' of nodes. This is extremely attractive to institutional investors who are averse to 'black box清算'. Additionally, its ZK-Orderbook design is naturally resistant to MEV, protecting the strategy privacy of high-frequency traders.

QWhat are the three key future evolution directions for Perp DEXs mentioned in the article's conclusion?

AThe three key future evolution directions are: 1) Intent-centric trading体验, where users express an intent and the system finds the optimal execution path across chains. 2) The爆发 of AI Agents, where a significant portion of open interest will be driven by AI, requiring DEXs to provide greater computational power and lower data latency. 3) The evolution of pricing models, moving beyond simple AMM or order books to incorporate complex dynamic risk engines that automatically adjust system parameters based on real-time volatility and position deviation for superior robustness.

İlgili Okumalar

Ethereum Q1 2026 Report: Fees Decline, Users and Transaction Volume Hit New Highs

Ethereum Q1 2026 Report: Fees Down, Users & Transactions Hit New Highs Token Terminal's Q1 2026 report on Ethereum presents a pivotal development: the network achieved record highs in monthly active users (13.2M, +85.9% YoY), total transactions (200.4M, +81.5% YoY), and throughput (25.78 TPS), while transaction fees on the mainnet plummeted by 47.9% quarter-over-quarter. This shift is attributed to the network's strategic move into a "low fees for scale" phase, exemplified by the Fusaka upgrade which increased data capacity and lowered block space costs, releasing pent-up demand (a manifestation of Jevons's Paradox). The report highlights a core narrative shift for Ethereum: from a DeFi-centric blockchain to a global financial settlement layer. It maintains a dominant position in tokenized assets, holding majority market shares among top chains in stablecoins (61.8%), tokenized funds (73.0%), and tokenized commodities (84.0%). Growth in tokenized funds (+73.1% YoY) and commodities (+325.9% YoY) was particularly strong, driven by institutions like BlackRock and JPMorgan entering the space. Contrasting these usage gains, several USD-denominated value metrics declined in Q1: fully diluted market cap fell 30.3% QoQ, total value locked (TVL) dropped 11.0%, and ecosystem transaction volume decreased 24.0%. The report interprets this as Ethereum prioritizing long-term network expansion and cementing its role as the default settlement layer for finance over short-term fee capture. The commentary from Etherealize argues that, much like the early internet, Ethereum's open, permissionless model is poised to win over closed alternatives as institutional tokenization accelerates.

marsbit1 saat önce

Ethereum Q1 2026 Report: Fees Decline, Users and Transaction Volume Hit New Highs

marsbit1 saat önce

He Just Raised 2.7 Billion, and Li Fei-Fei Also Invested

Pete Florence, a former senior research scientist at Google DeepMind and a key contributor to the Vision-Language-Action (VLA) model architecture, is deliberately distancing his startup, Generalist AI, from the trendy "world model" label. He argues that the industry should prioritize concrete goals over buzzwords. His goal is to create robots that can perform a vast range of unseen tasks with high speed and success rates, without needing task-specific training data. Recently, his company raised $400 million (¥2.7 billion) at a $2 billion valuation. Notable investors include NVIDIA's NVentures, Bezos Expeditions, NFDG, as well as Xiaomi co-founder Lin Bin, Zoom founder Eric Yuan, and renowned AI scientist Fei-Fei Li. Florence's approach stems from his academic background at MIT under Professor Russ Tedrake, focusing on understanding the physical world. After joining DeepMind, he developed models like Transporter Network and co-created the VLA framework. He left in 2025 to found Generalist AI. The company has launched two models: GEN-0, which demonstrated that scaling laws apply to physical motion, and GEN-1. GEN-1 was trained on over 500,000 hours of physical interaction data collected via a specialized wearable device. It achieves a 99% success rate on precise mechanical tasks like folding boxes and maintains performance three times faster than its predecessor. Florence believes GEN-1 is reaching a commercial utility threshold similar to the GPT-3 inflection point. The substantial funding round, following GEN-1's release, signifies strong investor confidence in Generalist AI's practical, goal-driven path to creating versatile, useful robots, regardless of the "world model" terminology.

marsbit1 saat önce

He Just Raised 2.7 Billion, and Li Fei-Fei Also Invested

marsbit1 saat önce

Two Legends Lost in Three Days: Is Google's AI Talent Dam Cracking?

In three days, Google lost two AI legends. On June 18, Noam Shazeer, co-author of the seminal "Attention is All You Need" paper and Gemini co-lead, left for OpenAI. Just 48 hours later, John Jumper, 2024 Nobel laureate and AlphaFold lead, departed DeepMind for Anthropic. This follows Andrej Karpathy joining Anthropic in May. These moves highlight a structural trend: top AI talent is concentrating at mission-driven, pre-IPO firms like OpenAI and Anthropic, while Google becomes a primary source. The exodus stems from a core mission mismatch. Google's ad-centric model often subordinates AI research to product and revenue goals, creating friction for pioneers like Shazeer, who returned in 2024 only to leave again. In contrast, OpenAI and Anthropic offer singular focus on pushing AI boundaries, whether towards AGI or safety-aligned models, which deeply appeals to top researchers like Jumper. Financial incentives amplify the pull. With both OpenAI and Anthropic nearing IPO, employees stand to gain immensely from equity, an upside Google's mature stock cannot match. Furthermore, the 2023 merger of Google Brain and DeepMind, intended to consolidate strength, has instead created cultural tension and slowed the path from research to product, as evidenced by Gemini's pace. This talent redistribution is reshaping the AI landscape. While Google retains vast data and compute resources, its true crisis is the quiet, continuous loss of the people who define the field's future. The real moat in AI is not infrastructure, but the concentration of brilliant minds—a battle Google is currently losing.

marsbit3 saat önce

Two Legends Lost in Three Days: Is Google's AI Talent Dam Cracking?

marsbit3 saat önce

Behind the AI Report Card, Lies a Chinese 'Exam Setter'

Beyond the familiar performance charts like MMLU-Pro and MMMU, which major AI models strive to ace, stands a key "examiner": Chinese-Canadian researcher Wenhu Chen. An assistant professor at the University of Waterloo and founder of TIGERLab, Chen addresses the crucial need for more rigorous AI evaluation. As models like GPT-4 began scoring near-perfect results on older benchmarks like MMLU, it became difficult to distinguish their true capabilities. In response, Chen introduced MMLU-Pro in 2024, featuring harder, more reasoning-focused questions with more answer choices, successfully reintroducing meaningful performance gaps. His work extends to multi-modal evaluation with MMMU and its enhanced version, MMMU-Pro. These benchmarks test a model's ability to understand and reason with complex information from images, charts, and text across diverse academic subjects, exposing the significant challenges even top models face in genuine comprehension. Chen's background in complex QA, table reasoning, and his experience at Google DeepMind on projects like Gemini inform his approach. He understands that effective benchmarks must anticipate how models might "cheat" by memorizing data or avoiding visual analysis. His lab also actively researches video understanding and generation models (e.g., UniVideo, Vamba), ensuring his evaluation work is grounded in practical model-building challenges. Now at Meta's Super Intelligence Lab, Chen continues his focus on multi-modal data and evaluation, representing the deep yet often unseen contributions of Chinese talent in shaping the fundamental tools of the AI industry.

marsbit3 saat önce

Behind the AI Report Card, Lies a Chinese 'Exam Setter'

marsbit3 saat önce

İşlemler

Spot
Futures
活动图片