Part Three丨Depth Analysis of Perp DEX: Hyperliquid, Aster, Lighter, edgeX

Lighter - Don’t trust—verify

This generation of Perp DEX has fundamentally differed from the previous generation of Perp DEX such as “GMX, DYDX”. Since it is still a Perp DEX, the title will continue to discuss it as a Perp DEX.

Lighter is a Perp DEX invested by the well-known American venture capital firm a16z, based on Ethereum's ZK Rollup, and it also uses an order book (CLOB) model.

During the AMA at the end of August this year, the team mentioned some future plans for Lighter. For example, the launch of RWA derivatives and Pre-launch perps (these two developments have lower volumes than spot trading and are planned to be launched almost simultaneously). Season one of the points competition (private testing phase) is about to end, and the distribution rules for points during this phase are intentionally not fully disclosed to prevent score manipulation. Season two (after public testing) will provide clearer guidance. Additionally, when entering the next phase, there will be a scheduled disclosure of some well-known investors and their contributions. The most important point is that the TGE is still progressing normally and is expected to take place in Q4 of this year.

Today (2025.10.3), after 8 months of private testing, the Lighter public mainnet is officially launched. However, there are no signs of RWA perps, Pre-launch perps, or Spot being released, and it's unclear if the TGE will be delayed. A wild guess: Spot and TGE might launch simultaneously, while RWA perps and Pre-launch perps will be released during the public testing period.

The above image shows the 24-hour data of Perp DEX on October 1, 2025. Hyperliquid, Aster, and Lighter occupy the top three positions in terms of trading volume and OI data. According to the Volume/OI ratio, Hyperliquid is less than 1 (a large number of users hold positions overnight, indicating a healthier market), with a significant real trading volume and user proportion; Lighter is around 6, which is at a medium level; Aster is about 40, the highest, indicating that the current volume manipulation in Aster is severe as the new season kicks off.

The first two TGEs were very successful, whether Lighter will be the next “epic big hair” after Hyperliquid and Aster is still uncertain, but the community's expectations are very high.

Lighter user growth flywheel and points incentive mechanism design

The growth logic of Lighter revolves around two core aspects: “invitation-based cold start + points-driven fission”.

During the cold start phase (lasting 8 months), Lighter used a closed testing strategy where only users holding an invitation code could register (public testing just opened today, and an additional 3 invitation codes can be obtained if the weekly points exceed 50). The difficulty of obtaining invitation codes is relatively high, and existing users must accumulate a certain trading volume or points to unlock a small number of invitation codes. For each successful invitation of a new user, the inviter receives an additional 10% points reward from the invited user. This design increases the cost of “batch registering small accounts,” effectively suppressing witch attacks, and creates a market thirst effect through the scarcity of invitation codes. There has even been off-market trading of invitation codes and points within the community, further reinforcing users' FOMO psychology. A fixed distribution of 250,000 points is issued weekly, with limited points and no inflation.

Cold start, fission, prevention of witch attacks, and airdrop incentives are organically combined to build a self-reinforcing growth flywheel: Expected airdrop → Brush trading to accumulate points → Unlock invitation code → Fission propagation → Community expansion → Increase in point value → Attract users again.

Some scoring strategies based on the tips provided by the officials.

  1. Small capital (< $10k) scoring strategy •Asset selection: Choose trading assets with low open interest (OI) and low trading volume (as advised by the official), making it easier for the scoring model to “see” the trading volume/positions without pushing up the slippage.
  2. Ordering method: primarily use small, multiple limit orders to reduce impact costs; avoid “unsystematic order waterfall.” 2. Risk: Low liquidity means higher slippage and liquidation risk - it's best to keep leverage ≤ 5x, maintain a margin reserve of about 2 times as a buffer, strictly set take profit and stop loss, and strictly control liquidation (1% nominal principal liquidation fee is very “painful”).
  3. Sub-accounts: Used to isolate strategies/risk control (rather than duplicate multi-account behavior). Official explanation: Sub-accounts are included in statistics, but the witch judgment will be reset to zero, and do not abuse multiple accounts.
  4. Medium/Large Capital (≥ $10k–$1m) and Semi-Automatic/Quantitative
    1. API Trading: Generate a key, after receiving the API role in Discord, use low-impact execution methods such as TWAP/VWAP; • Control the order-fill ratio and cancellation rate (for example, a cancellation rate ≤70% as an internal red line) to avoid being classified by the system as wash trading/HFT abuse.
    2. Fees and Identity: If the trading style is close to market making/HFT, calculate the marginal returns of the strategy based on costs of 0.02%/0.002%; accounts that are more medium-term/trend-oriented should try to maintain a retail profile (0% transaction fee) characterized by low frequency, low withdrawal, high transaction volume, and long holding periods.
    3. Hedging and Risk: Directional hedging/basis trading can be conducted on other exchanges to achieve “normal trading” rather than self-trading; any acts of matched orders/same account counterparties may be considered wash trading.
  5. User Public Pools' Points and Capital Efficiency
    1. Passive Participants: If DeFi Farmers want to “invest to earn points” (the latest policy requires obtaining 50 points first to have a 25% quota), they should prioritize pools created by users (as the pools operated by the protocol, LLP, do not earn points). Selection criteria: historical drawdown, net value curve, maximum position, win rate over the past 30 days, and average holding duration.
  6. Active Operators: Building a user pool can simultaneously expand the capital scale and the source of points, but it is essential to have self-owned funds for co-investment, publicly disclose strategy rules and risk control lines, and establish credibility; avoid excessive risk-taking for “point chasing.”

Based on my own experience of trading for several weeks, I would like to summarize some guesses about real trading points (anti-rebate volume). Here are three points (not necessarily correct, welcome to discuss):

  1. Formula: f = ( actual transaction amount × maker quality × OI‑hours × price impact negative reward ), if any item is 0, all are 0.

  2. Multiplier: Market making has weight in key depth zones (such as within 1bp); the longer the net position is held, the higher the weight (encourages real risk tolerance).

  3. Exclusion: Short-term mutual transactions with the same account/same IP/same device, extremely high cancellation ratio, net OI ≈ 0 “self-transaction” ring

Try to avoid two extremes: one is high trading volume X OI-hours with quick openings and closings, and the other is low trading volume X high OI-hours for hedging positions. The final result of the multiplication is not very large (just a guess).

Public Fund Pool

The public fund pool includes the protocol public fund pool and the user protocol public fund pool, nothing special, similar to Hyperliqyid.

In the LLP section, Lighter implements a 0 fee policy for regular traders (both Maker and Taker fees are 0), aiming to profit through high-frequency traders. As of today, the public test begins, and regular users can still trade completely for free, with only the funding rate settled between long and short positions, and the platform does not charge any additional fees.

In this model, the profits of the LLP mainly come from market-making profits and liquidation gains: when the LP fund pool serves as the counterparty, if most traders incur losses, the corresponding gains and losses will enter the LLP. The current LLP APR is approximately 60% (for example, in September, there were 21 days with positive returns and 9 days with negative returns).

At the same time, LLP can charge a certain fee or retain a margin during liquidation. Lighter's liquidation mechanism stipulates that during certain forced liquidations, orders may be matched at a “0 price”; if a better price is achieved, no more than 1% of the liquidation fee will be injected into LLP. Therefore, whenever forced position reduction or liquidation occurs, LLP plays the role of an insurance fund, earning income from liquidation fees. Additionally, LLP, as a counterparty, also earns funding fees (the funding fees paid by the counterparty flow to LP) and spreads. Since transaction fees are zero in the early stages, Lighter platform has limited stable income sources, and LLP's earnings largely depend on market fluctuations and traders' profit and loss situations. Overall, Lighter LLP currently generates revenue for LP through counterparty profit and loss, funding fees, and liquidation fees, and will introduce a portion of fee sharing in the future.

In addition, after the public test starts today, the threshold for the public fund pool LLP has increased from “0 points can deposit 25%” to “must first obtain 50 points to deposit 25%”, which maintains a high APR to a certain extent, protecting the interests of early users from being impacted and diluted.

The user public fund pool portion is similar to Hyperliquid, resembling a secondary fund and copy trading system, where fund managers/copy traders receive a commission of 20-30%. The difference is that participants in the user public fund pool can earn points, which LLP cannot, but the risks are also higher than those associated with LLP, so caution is required.

Due to the lack of token issuance, Lighter's current user retention mainly relies on incentive points and community operations, with “zero transaction fees” still being one of Lighter's biggest selling points. As someone who is optimistic about Lighter and intends to hold Lighter tokens for the long term, here are some suggestions for Lighter in the post-TGE era. First, take a cue from Aster, where after the first round of airdrop hype, the second phase of activities was launched immediately, issuing a secondary airdrop, successfully retaining some users into the second phase. Lighter could also consider reserving a portion of tokens for rewarding users who continue to trade after the token issuance (for example, distributing based on monthly or quarterly trading volume and active days), to smooth the transition during the inactivity period following the airdrop. Second, simultaneously with the TGE, introduce a VIP tier and fee discount system. Drawing from Hyperliquid (which launched a bit late), establish user tiers based on token holdings or trading volume. For example, holding a certain amount of platform tokens or reaching a trading threshold in the past 30 days could allow users to upgrade their VIP status and enjoy benefits such as fee discounts, priority customer service, and testing qualifications. This approach not only encourages users to hold their tokens (not to sell them) but also motivates high-end users to continue trading to maintain their discounts, forming a retention loop.

Overall, Lighter technically introduces ZK to ensure fairness, offers a zero-fee product that is user-friendly, and is poised to gain momentum through its points mechanism. It will definitely have a place in the Perp dex space. Also, as mentioned in the AMA at the end of August, Lighter will rhythmically disclose some well-known investors in the next phase. Today, the public testing officially begins, and the official Twitter account has received endorsements from multiple Ethereum ecosystem founders like EigenLayer. Additionally, there are reports that Founders Fund, Arthur Hayes, and others have participated in the investment.

Finally, let's talk about Lighter and the “Ethereum Finale”.

Lighter does not build its own L1 and complete protocol like HyperLiquid; instead, it will serve as a native composable Layer 2 on Ethereum— for example, by tokenizing LLP to the mainnet and collaborating with protocols like Aave, enjoying a low-cost, low-latency, and verifiable execution environment. The ecosystem follows the “Ethereum composability” route to connect faster to mainstream DeFi, rather than a closed-loop self-built approach.

Lighter = “Verifiable Exchange” at the Ethereum endpoint. It incorporates the two most sensitive aspects of exchanges—price-time priority matching and settlement—into ZK proofs, while entrusting asset custody, state, and data availability to Ethereum; as 4844 → Danksharding, (e) PBS, and (potential) native Rollups gradually roll out, Lighter's advantages in cost, anti-censorship, and trust minimization will increasingly resemble “protocol-level features.” This is its structurally aligned relationship with the Ethereum endpoint.

If Aster is backed by Binance, then Lighter is backed by Ethereum.

For details on edgeX, see the following article: “In-Depth Analysis of Perp DEX: Hyperliquid, Aster, Lighter, edgeX (4)”

(The above is just a personal opinion and not investment advice. Please feel free to point out any mistakes.)

PERP-7.19%
HYPE-0.07%
ASTER-2.23%
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
0/400
No comments
Trade Crypto Anywhere Anytime
qrCode
Scan to download Gate App
Community
English
  • 简体中文
  • English
  • Tiếng Việt
  • 繁體中文
  • Español
  • Русский
  • Français (Afrique)
  • Português (Portugal)
  • Bahasa Indonesia
  • 日本語
  • بالعربية
  • Українська
  • Português (Brasil)