FirstVIP Research Report: LayerZero, An Omnichain Interoperability Protocol

Project Overview

LayerZero is an omnichain interoperability protocol focused on data messaging between chains. In the industry, this type of “bridge” is often referred to as “Arbitrary Messaging Bridges (AMBs),” which allow for the transfer of any data, including tokens, chain states, contract calls, NFTs, or governance votes, from Chain A to Chain B.

In the field of cross-chain bridges, we have previously seen primarily “asset cross-chain” projects. However, we are now seeing some projects gradually shifting towards exploring the field of data transmission. LayerZero is one of the pioneers in this area.

The highlights of the LayerZero project are:

1) The LayerZero Foundation currently holds a total asset value of $261 million, providing sufficient funds for the long-term development and operation of the project.

2) In terms of product design, LayerZero differs from traditional cross-chain bridges on the market by utilizing oracle networks instead of continuous streaming for cross-chain transfers. By outsourcing the burden of verifying on-chain information to third-party oracles, the protocol becomes more lightweight and cost-effective to operate.

3) LayerZero’s innovative technology, combined with faster deployment speed and certain cost advantages, as well as the early support from renowned VCs and influential community KOLs, has led to rapid expansion of the LayerZero ecosystem in approximately one year. The project has achieved significant milestones in the DeFi, NFT, and stablecoin sectors. Currently, there are more than 50 projects (including projects not yet officially launched/online) integrating or using LayerZero technology.

4) The number of well-developed and unattacked Arbitrary Messaging Bridges (AMBs) projects on the market is still relatively small, giving LayerZero a certain first-mover advantage.

The risks of this project are:

1) The security of LayerZero has not been fully validated, and the trust assumptions between oracles and relayers need further consideration. The security vulnerabilities behind the relaying mechanism should also be carefully monitored. However, on the other hand, LayerZero’s security theoretically does not fall below the trust assumptions of oracles, which is convincing. The key point may lie in achieving decentralized relaying.

2) LayerZero’s economic model has not been released yet. In the field of cross-chain bridges, most project tokens have generally shown weak value-capturing abilities. The future economic model of LayerZero remains to be observed.

Overall, although LayerZero still faces some challenges, its fundamentals are generally strong, making it worth paying attention to.

Note: The final “Focus” / “Not Focus” determined by FirstVIP is the result of a comprehensive analysis of the current fundamentals of the project based on the FirstVIP project evaluation framework, not a prediction of the future price movement of the project’s token. There are many factors that influence token prices, and project fundamentals are not the only factor. Therefore, it should not be assumed that a project will definitely experience a price drop just because it is determined as “Not Focus” in the research report. In addition, the development of blockchain projects is dynamic. If a project determined as “Not Focus” undergoes significant positive changes in its fundamentals, we may adjust it to “Focus.” Similarly, if a project determined as “Focus” undergoes significant negative changes, we will issue warnings to all members and may adjust it to “Not Focus.”

1. Overview

1.1 Project Introduction

LayerZero is an interoperability protocol designed for lightweight information transfer across different chains.

It is important to note that LayerZero focuses only on message passing between chains and is capable of sending messages to any smart contract on any supported chain. It acts as a messaging layer for communication between smart contracts across blockchains and does not handle cross-chain asset transfers.

1.2 Basic information

2. Detailed explanation of the project

2.1 Team

LayerZero Labs Canada Inc. (Company Number: 1355847-9) was registered in Canada under the Canada Business Corporations Act on November 30, 2021. Caleb Banister, Ryan Zarick and Bryan Pellegrino are listed as company directors[1].

According to LinkedIn[2], currently, LayerZero has 29 members. The details of the core members are as follows:

Caleb Banister, co-founder of LayerZero Labs and Stargate Finance, graduated from the University of New Hampshire in the United States in 2010. From 2005.06 to 2010.12, he worked as a software developer at the UNH Interoperability Lab. From 2010.09 to 2021.02, he was a co-founder of Coder Den, a software consulting company. From 2018.03 to 2021.02, he was a co-founder of 80Trill, a crypto company specializing in writing and auditing smart contracts for blockchain-related projects. From 2019.06 to 2021.02, he was a co-founder of Minimal AI, an ML/AI company. Since 2021.02, he has founded LayerZero.

BBryan Pellegrino, co-founder and CEO of LayerZero Labs, graduated from the University of New Hampshire in 2008. From October 2010 to January 2013, he served as co-founder and COO of Coder Den. From June 2011 to January 2013, he was the CEO of BuzzDraft (acquired in 2013). From October 2017 to August 2019, he was a co-founder of OpenToken. Since June 2016, he has been the Chief Engineer at Rho AI. He founded LayerZero in 2021. Prior to founding LayerZero, Pellegrino was a professional poker player and successfully sold a set of machine learning tools he developed to a Major League Baseball (MLB) team. He has also published reports in the field of artificial intelligence. Mario Gabriele, a Generalist, conducted an interview with Pellegrino, and those interested in his background can refer to the following link.

Ryan Zarick, Co-founder and CTO of LayerZero Labs, graduated from the University of New Hampshire in 2011. From 2006.08 to 2011.05, he worked as a software developer and graduate assistant at UNH Interoperability Lab. From 2011.11 to 2013.03, he served as the CTO of BuzzDraft. From 2010.09 to 2020.13, he co-founded Coder Den. From 2018.01 to 2020.03, he co-founded 80Trill. From 2019.06 to 2021.01, he co-founded Minimal AI. In 2021, he founded LayerZero and became the CTO.

Judging from the resumes of the three co-founders of LayerZero Labs, there is a high level of overlap, indicating a long-standing cooperative relationship and a well-coordinated team. All three individuals have years of development experience or successful entrepreneurial experience.

0xMaki[3], a former founding member and core contributor of SushiSwap, has now joined LayerZero Labs full-time. 0xMaki played a key role in the early marketing of SushiSwap and became the project leader after Chef Nomi exited. During his tenure, 0xMaki was mainly responsible for determining the daily operations, business development strategy and overall development of SushiSwap. In addition, Sushi’s cross-chain Swap project SushiXSwap was completed under the leadership of 0xmaki, adding application scenarios to the Sushi and LayerZero protocols.

2.2 Funding

Table 2-1 LayerZero Financing Situation

In addition, due to the impact of the FTX’s bankruptcy incident in early November 2022, on November 11, 2022, LayerZero officially announced that it has repurchased 100% of the equity, token rights, and any other agreements from FTX/FTX Ventures/Alameda Research. At that time, the total value of assets held by the foundation amounted to $134 million [8] (the $10.7 million held by the team on the FTX exchange is not included in the calculation mentioned above). Therefore, it can also be seen that the third round of financing for LayerZero has not been completed.

From the table above, we can see that LayerZero, as a star project, has been favored by major capital sources from the very beginning. The known total amount of funding raised so far has reached $261 million. Overall, LayerZero currently has abundant funds, sufficient for the long-term development and operation of the project.

2.3 Code

Figure 2-1 LayerZero code base situation[9]

As shown in Figure 2-1 above, the LayerZero code base has been updated since March 2019. Overall, LayerZero has accumulated 6,415 code submissions, and a total of 116 developers have become Git/Issue authors and review submitters on LayerZero Github.

Based on the progress disclosed by LayerZero in September 2022 [10], the testnet of LayerZero has deployed over 7000 active contracts, indicating a very good adoption rate.

In addition, the LayerZero code base has completed a total of 4 audits conducted by Zellic, Ackee and SlowMist (SlowMist). The specific audit reports can be found through this link.

To sum up, in the past three years, the LayerZero project has had good code changes, sufficient developers, and several important code bases have been updated frequently.

2.4 Technology

First, we need to clarify a common misconception: LayerZero is an omnichain interoperability protocol that focuses only on message transmission between chains. It can send messages to any smart contract on any supported chain, serving as a message transport layer for smart contract communication between blockchains, but it is not responsible for cross-chain asset transfers.

2.4.1LayerZero Framework

According to the LayerZero whitepaper [11], the protocol core consists of three components: Endpoint, Oracle, and Relayer.

1) EndpointIt is a facility that directly interacts with users or applications, or it can also be regarded as a series of smart contracts that process logic. These endpoints handle message transmission, validation, and reception. Their purpose is to ensure efficient delivery when users send messages using the protocol.

In the LayerZero protocol, each chain needs to deploy a LayerZero Endpoint. Endpoint can be called and used by other apps in the same chain, and is responsible for sending information to external links. For example: if a Dapp wants to transfer information from Chain A to Chain B, it must first call the Endpoint of Chain A and submit the information that needs to be sent.

Each LayerZero Endpoint is divided into 4 modules: Communicator, Validator, Network, and Libraries. Communicators, validators, and network modules make up the core functionality of Endpoint, and these modules act similarly to a traditional network stack. Messages are sent down the stack on the sender (communicator), verified by a validator before being passed to the network, and then sent up the stack on the receiver.

Each new chain supported by LayerZero is added as an additional library. These libraries are auxiliary smart contracts that define how specific communications for each chain are handled. Each chain in the LayerZero network has an associated library, and each endpoint includes a copy of each library.

Before introducing oracles and relays, we need to clarify a concept first. First, in order to verify a block on the chain, we need two pieces of information: 1) block header, which contains the Receipts Root[12]；2）Proof of transaction, that is, the Merkel-Patricia proof on EVM[13]。

LayerZero separates these two parts in the following ways: 1) The oracle forwards the block header——Any selected oracle; 2) Relayer forwards transaction proof.

2) For LayerZero, the oracle is an external component, i.e., a third-party service independent of the LayerZero protocol. The main value provided by the oracle is to send block headers to another chain, so that the validity of transactions on the source chain can be verified on the target chain.

3) The relayer is an off-chain service that retrieves transaction proofs from the source chain and then transfers them to the target chain. LayerZero believes that in order to ensure that transactions can be effectively delivered, the oracle and relayer must be independent of each other.

Currently, the most common way for one chain to communicate with another chain without trust is to continuously transmit the block headers of Chain A to Chain B. For example, the Relay relays the BTC block headers through a third party, providing a trusted BTC data source for cross-chain applications on Ethereum, enabling the circulation of value between BTC and Ethereum. In this case, the cross-chain bridge contract is basically a light client. This method of transmitting information is the most secure, but the problem is that the cost of writing to the blockchain is very high, so continuously transmitting these block headers is very expensive.

The biggest improvement of LayerZero is that it chooses an oracle network to replace this continuous streaming.

Currently, according to the LayerZero official website documentation and team disclosure, Chainlink and TSS Oracle are the oracles configured on the testnet. The current oracles are not decentralized and have not been tested in real-world scenarios, which means there is a risk of being hacked. According to the official description, after the completion of LayerZero’s testing, more oracles will be disclosed.

LayerZero uses Chainlink as its oracle, which will have several benefits:

1) Outsourcing the functionality of verifying information eliminates the need to run nodes on the linked chain. Oracles only allow the transfer of block headers to the target chain once, reducing operating costs;

2）LayerZero uses oracles and relayers to transmit messages between endpoints on different chains. By streaming block headers on-demand through oracles, LayerZero achieves the desired synchronization state with more efficient off-chain entities. The block headers submitted by oracles are cross-validated with the transaction credentials submitted by relays. Only when oracles and relayers collaborate will the system fail, ensuring security no less than that of the oracles;

3）Neither the Relayer nor the Oracle form any consensus or verification, they just transmit information.Since all verification is done on the respective source and destination chains,The speed and throughput limits depend entirely on the properties of the two transaction chains.

However, there are also drawbacks: LayerZero outsources the task of validating on-chain information to a third party, such as the upcoming use of Chainlink. This is not to say that Chainlink is bad, but rather that LayerZero introduces security assumptions that the protocol cannot control. In the long run, shifting the workload of critical tasks to third parties increases other risks and potential uncertainties.

2.4.2 Safety

•In LayerZero, there is an important trust assumption that oracles and relayers need to operate independently from each other.

To ensure the effective transmission of information, if there is any dispute in the information exchange between relayers or oracles, the smart contract will pause and not submit the information to the target chain. This means that the system will only crash when the oracle and relayer collude, ensuring security not lower than that of the oracle.

Although in LayerZero, the protocol allows each Dapp’s development team to modify the oracle/relayer code provided by LayerZero, and graft it onto their own servers or validator networks to use their own oracles for price feeding, or run their own relayer to ensure that the oracle does not collude with the relayer (LayerZero has also previously suggested that relayers need to be more decentralized).

However, the current situation is that although everyone knows that “decentralization” is better, most Dapps, due to cost, operation, user experience considerations, and the notion that “Chainlink is good enough”, prefer Chainlink as their oracle of choice. Similarly, most Dapps would directly choose LayerZero’s relayer. It’s similar to how there are hardly any users running their own nodes for trading, as people rely on centralized service providers like Infura and Alchemy.

In this case, if a relayer exhibits malicious behavior (being hacked or not performing as expected), the Chainlink oracle will intercept and prevent any significant loss on the original chain. The advantages of choosing Chainlink are undeniable, but if we assume that Chainlink can be an effective and realistic alternative to achieve both functionalities (oracle and relayer), then the trust assumption of LayerZero becomes questionable.

The above viewpoint was inspired by Pickle and Aylo’s article “Layer 0 Wars: LayerZero vs Chainlink’s CCIP”. Interested readers can refer to the original article for further reading.

•The security of the Chainlink oracle has been validated by the market, and the key to the security features within the LayerZero protocol lies in the relayers.

In April 2022, the LayerZero team introduced a method to ensure protocol security, called “Pre-Crime.” Currently, there is limited public information about Pre-Crime, and the blog post only provides an overview of its underlying operation. In summary, the Pre-Crime model allows User Applications (UAs) to define a set of specific assertions, which the relayers must validate. If the assertions fail, the relayer will not relay the transaction. By introducing Pre-Crime, the relayers can prevent hacker attacks before they occur[14].

Currently, the corresponding code repository for “Pre-Crime” has not been open-sourced. However, the LayerZero team has launched a private Pre-Crime beta version with multiple teams. The release date of the official version is yet to be disclosed, and its effectiveness still needs to be verified through practice.

•Security risks behind the relaying mechanism[15]

Previously, on March 28th, LayerZero updated the verification contract used for cross-chain transactions without making any public announcements. The Cobo Security Team discovered that this update was a fix for a significant security vulnerability by comparing the code of the original validation contract (MPTValidator) and the new validation contract (MPTValidatorV2).

The code for this vulnerability is the most critical part of the MPT transaction validation in the LayerZero protocol and serves as the foundation for the normal operation of the entire LayerZero and upper-layer protocols. If not detected in a timely manner, the most severe consequence that can occur, even with full trust in the LayerZero oracle, is that the relayers can still attack the cross-chain protocol by forging receipt data, breaking the previous security assumptions of LayerZero.

Although LayerZero has fixed the current vulnerability, the possibility of other vulnerabilities cannot be ruled out. This incident has also raised concerns in the community about the security of the relaying mechanism behind LayerZero.

In summary, although LayerZero has grown to a considerable size, the security behind its protocol has not yet been fully verified.

2.4.3 Execution Process

Figure 2-2 Communication flow in LayerZero Cross-chain Transactions

The specific execution process of LayerZero is as follows:

• When User Application[16] passes a cross-chain message (for example, from Chain A to Chain B), it first needs to call the LayerZero Endpoint smart contract.

• The message enters the Endpoint of Chain A, and then this endpoint packages the message (transaction proof and block header) and information to Chain B (target chain) to the oracle and relayer (both entities are independent and off-chain).

• The oracle reads and confirms the block header. After the oracle determines that the block has been confirmed several times on Chain A, it sends the block header to the Endpoint on Chain B. At the same time, the relayer submits the corresponding transaction proof.

• After the target chain successfully verifies the block header and transaction proof, the message is forwarded to the target chain, completing the cross-chain communication.

Note: In order to make the above process easier to understand, the editor has simplified some details, such as the endpoints (communicator, validator and network), but the essential logic remains unchanged.

From the above process, it is easy to see that LayerZero is only responsible for message transmission, similar to A having a message that needs to be transmitted to B, so A calls B and tells them the content of the message, B picks up the phone, receives the message, and the process ends. This is a very simple logic. So how are cross-chain assets transferred?

Firstly, each chain needs to deploy a LayerZero Endpoint to send and receive information. The liquidity of asset transactions is balanced by DApps such as DEX that integrate LayerZero’s functionality at various endpoints.

Currently, Stargate Finance provides this balancing ability for LayerZero, and Stargate’s Delta (Δ) algorithm ensures that cross-chain liquidity remains balanced and available (for more details, refer to the report on Stargate Finance previously published by this translation).

In short, LayerZero is only responsible for addressing communication issues between chains, and other additional functionalities/problems are resolved by the applications that integrate LayerZero on their own.

2.5 Ecosystem

LayerZero is an Omnichain interoperability protocol. As a hub for information exchange between chains, LayerZero can do more than just cross-chain asset transfers. After achieving cross-chain message transmission, LayerZero can also enable cross-chain state sharing, lending, governance, and more.

Furthermore, unlike traditional cross-chain bridge models currently on the market, LayerZero does not require running nodes on each connected chain to monitor the state of the source chain. Instead, the role of validators is taken over by oracles. One obvious benefit is that there is no need to deploy a new node on every new chain. Starting from this point, LayerZero can integrate new chains into the network faster and at a lower cost. As of November 11, 2022, LayerZero has already supported a total of 13 chains, including Ethereum, BNB Chain, Avalanche, Aptos, Polygon, Arbitrum, Optimism, Fantom, and others.

LayerZero’s innovative technology, coupled with faster deployment speed and certain cost advantages, as well as the promotion of early famous VCs and influential KOLs in the community, have enabled the LayerZero ecosystem to expand rapidly in just a year or so, andIt has made great achievements in DeFi, NFT and stable currency.Up to now, there have been50+ items(Including projects that have not been officially launched/online) Integrated or using LayerZero technology. The details are as follows (only some are listed):

Figure 2-3 List of LayerZero ecological projects

Note: The above image is compiled and summarized by @LayerZeroHub (unofficial). If you want to track LayerZero’s ecological projects in the future, you can also follow the list maintained by Luke (Twitter ID: @0x4C756B65) on Twitter.

1) DeFi Field

Table 2-2 LayerZero Ecological DeFi Collaboration Projects

2) Stablecoin Field

Table 2-3 LayerZero Ecological Stablecoin Collaboration Projects

3) NFT Field

Table 2-4 LayerZero Ecological NFT Field Collaboration Projects

Combining the information from Figure 2-2 and Tables 2-1 to 2-3, we can see that the ecosystem of LayerZero has developed to a considerable scale. From the blue-chip DEXs like Sushi and PancakeSwap to the currently popular Radiant Capital, all of them are using LayerZero’s Stargate for cross-chain DEX development. In the stablecoin field, both USDC and agEUR are supported by LayerZero technology for cross-chain interoperability of their respective stablecoins, upgrading them to multi-chain native assets. In the NFT field, although the demand for multi-chain NFTs is not yet significant, we have also seen attempts in the direction of multi-chain NFTs with projects like Gh0stly Gh0sts and tofuNFT. Additionally, LayerZero recently launched its official browser, LayerZero Scan, where cross-chain transactions can be bound to a database, allowing users and developers to extract transaction status, state, and time.

Through the measures taken both internally and externally by LayerZero, its concept of omnichain may further develop in the future.

In Summary:

LayerZero is an omnichain interoperability protocol designed for transferring lightweight information across chains. The overall architecture is reasonable and eliminates the need to run nodes on connected chains. By relying on oracles and relays, communication on different chains is Transfer messages between endpoints. Although the security has not been fully verified by the market, the protocol is theoretically no less secure than the oracle (Chainlink) and has certain guarantees.

The current value of assets held by the LayerZero Foundation totals US$261 million, and its treasury is very abundant. The code change of the LayerZero project is in good condition, and the ecosystem has expanded rapidly in just about a year. It is currently one of the fastest-growing projects in the cross-chain field.

3.Development

3.1 History

Table 3-1 Major events of LayerZero

3.2 Current situation

3.2.1 Network usage

Figure 3-1 LayerZero daily transaction number[17]

Figure 3-2 LayerZero cumulative number of transactions

From Figure 3-1 and Figure 3-2, the usage of the LayerZero network can be clearly seen. Over the past year, it has shown a steady upward trend. Especially in March 2023, when Arbitrum announced the airdrop of governance token ARB to its community members, the “airdrop frenzy” in the community reached an unprecedented high, leading to a significant increase in the usage of both the unreleased LayerZero ecosystem and zk ecosystem. Although this phenomenon may not be sustained in the long term, this “airdrop expectation” indirectly allows more users to understand LayerZero, thus retaining a certain number of real users.

In addition, even if LayerZero’s March data is taken away, its network adoption rate has doubled from the end of 2022 to the beginning of March. At present, we can also see that many protocols based on LayerZero have begun to be implemented, and initial results have been achieved in ecological construction.

Figure 3-3 Ranking of cross-chain asset volume for cross-chain bridges [18]

In addition, according to the data interface of DeFiLlama (as shown in Figure 3-3), the current bridged asset volume of Stargate, a project under LayerZero, is ranked first among all cross-chain bridges (including official bridges of various public chains and layer 2 solutions). Judging from the volume alone, Stargate has become the leading project in the cross-chain bridge race.

Note: The trading volume and number of transactions of various cross-chain bridges displayed on the DeFiLlama data portal are currently fluctuating greatly. This data does not represent the long-term competitive advantage of each cross-chain bridge and is for reference only.

However, it should be noted that in terms of the number of transactions, Stargate far exceeds other cross-chain bridges, but the amount of cross-chain funds does not widen the gap. There are currently no sufficient signs of Stargate’s small-amount transaction operations. The experience is better.Therefore, it can be speculated that a considerable part of its transaction data may be due to LayerZero’s potential airdrop expectations.

Although various projects do not advocate for taking advantage of airdrops, from another perspective, it is precisely because of the potential airdrop expectations that LayerZero and Stargate have gained higher exposure and adoption. The revenue generated for the protocol is also substantial.

3.2.2 Revenue

Currently, there is no threshold for ecological applications to access LayerZero. LayerZero Labs’ current main income comes from transaction fees from Stargate Finance.

Transfers of non-STG tokens through the Stargate protocol will incur a 0.06% transfer fee. Of which 0.01% will be allocated to liquidity providers, 0.01% will be allocated to veSTG holders, and 0.04% will be allocated to the protocol’s treasury[19]。

Figure 3-4 Stargate’s monthly cross-chain amount[20]

According to the monthly transaction amount dashboard disclosed by Stargate, from the launch of Stargate in March 2022 to the present (April 7, 2023), the cumulative cross-chain transaction amount has reached $6,286,702,699, approximately $6.3 billion.

For ease of calculation, assuming that all of the 6.3 billion US dollars are non-STG token transfers, Stargate’s treasury would approximately receive a transaction fee income of $6.3 billion * 0.04% ≈ $2.52 million.

If we calculate based on the current scale, according to Token Terminal’s statistics, Stargate’s protocol revenue for the past 30 days is approximately $730,000. If the current scale is maintained, the future annual revenue will reach $8.89 million [21] (in an ideal scenario, this data is for reference only).

3.3 The future

LayerZero currently does not have a specific roadmap. The main focus at the moment is integrating and consolidating with some projects, while also expanding to more chains.

In Summary:

LayerZero has made rapid overall progress, with its network growth particularly evident in the past 2-3 months. However, the protocol has not yet disclosed a detailed roadmap.

4. Tokenomic Model

LayerZero Labs has not yet issued a token, but the team disclosed the $ZRO token information in the code of its official document. Combined with Figure 4-1 below, we can see that $ZRO may be used to pay gas fees on its chain in the future.

Figure 4-1 Layerzero official document[22]

In addition, the community has previously speculated that LayerZero will eventually become tokenized, because there is staking behavior during the operation of the LayerZero protocol, and those malicious behaviors from relayers will lose the pledged $ZRO tokens. But this is just speculation and has not been confirmed by the team.

5. Competition

LayerZero is an omnichain interoperability protocol designed for transmitting lightweight information across chains. It belongs to the cross-chain bridge track. If it is further broken down, it is a transmission bridge that supports data messages.

5.1 Industry Overview

In last year’s analysis of the cross-chain bridge track published in the FirstVIP, the editor categorized all cross-chain bridges as asset cross-chain for easier understanding and differentiation from Polkadot and Cosmos’ cross-chain. However, after a year of development, we are now seeing more and more “bridges” exploring the field of data transmission, not limited to basic asset cross-chain.

Now it is actually not difficult to distinguish the difference between cross-chain and cross-chain bridge between Polkadot and Cosmos. Polkadot and Cosmos are essentially chains that use a unified framework and have high interoperability. At the same time, they do not have any cross-chain advantages for chains outside the framework. The cross-chain between the two is more like Layer 0. Users need to implement cross-chain based on their own standards; as for the cross-chain bridge, the two chains can have different protocols, which solves the problem between different assets and different networks. Asset and data migration issues.

When we talked about the term “cross-chain bridge” before, in fact, it was often limited to the discussion of “asset cross-chain”, that is, a liquidity network or a trusted third party facilitates the transfer of token X from chain A to chain B.

However, asset cross-chain is just a relatively easy-to-implement function between chains. Cross-chain bridges can do more than just transfer tokens from chain A to chain B; it also involves communication at the data level. Continuing to use the definition of cross-chain bridges by Dmitriy Berenzon, a partner at 1kx research [23]: at an abstract level, people can define “bridges” as systems that transfer information between two or more blockchains. In this case, information can refer to assets, contract calls, identity proofs, or states.

In simple terms, a cross-chain bridge is a tool that connects chains, allowing tokens, assets, and data to be transferred from one chain to another. The two chains can have different protocols, rules, and governance models, and the bridge provides a secure way for them to communicate and interoperate.

There are currently three main types of cross-chain communication methods on the market: 1) asset exchange; 2) asset transfer; 3) general communication.

LayerZero, as a cross-chain bridge that supports data messaging, belongs to the third category mentioned above. In the competitive analysis section, we will focus on comparing bridges of this type. However, we will not extensively compare and analyze common asset cross-chain bridges currently available in the market in this chapter.

For this type of bridge that supports “data cross-chain”, many development teams dedicated to the cross-chain field have previously called it“Arbitrary Messaging Bridges (AMBs)”, the editor believes that its definition is more pertinent, so this statement will be used below. Simply translated it is:Any information transfer bridge，These bridges allow any data, including tokens, chain status, contract calls,NFT or governance voting, from the chainAtransfer to chainB[24].

5.2 Competitive Product Introduction

Currently, in addition to LayerZero, the Arbitrary Messaging Bridges (AMBs) that are highly discussed on the market include Wormhole, Nomad, Celer Inter-chain Message (IM), Multichain’s anyCall and Axelar, etc.

5.2.1 Axelar [25]

Axelar is a universal cross-chain basic protocol. It uses the Cross-chain Gateway Protocol (CGP) and the Cross-chain Transmission Protocol (CTP), and uses its own POS public chain as a witness chain to transfer information between any two public chains. Currently, it covers a total of 15 public chains including Ethereum, Cosmos, and Avalanche.

Execution Logic:

Axelar Network builds connections with external blockchains through its API. Essentially, it deploys smart contracts on other chains and monitors the relevant information of these contracts using lightweight node clients running on its own network’s validators. This information is then transmitted to the Axelar mainnet for voting and validation. Once validated, the information is written into blocks and the requirements of the smart contracts on the target chain are fulfilled. The following diagram illustrates the process:

Figure 5-1 Axelar network flow chart

The above diagram provides a simple description of the operation process of the Axelar network, but it is not detailed enough. Next, the editor will provide a more in-depth description of the relevant processes through examples:

Assumption: Axelar has established gateways (smart contracts) with source chain A and target chain B. A user from source chain A wants to transfer assets to target chain B. This is done through the following 5 steps:

1) The user initiates a cross-chain asset transfer request through the gateway of source chain A. The information is transmitted to the Axelar mainnet via the Cross-Chain Transfer Protocol (CTP).

2) The validators of the mainnet use threshold signature technology to generate a deposit address on source chain A. The user then deposits the required amount of assets into the corresponding address.

3) The validators running the source chain A light node client on the Axelar mainnet verify the block information of source chain A and confirm the information that assets have been deposited into the corresponding address.

4) The mainnet returns and conducts voting through the DPoS consensus mechanism. Once more than 90% of validators confirm the accuracy, the process proceeds.

5) The node runs the target chain B light node client and uses threshold signature technology to make a payment to the user’s target chain address.

The above 5 steps represent the process of asset cross-chain transfer in Axelar. As for data cross-chain transfer, the process is roughly similar, but it is more complex. The official information only discloses simple data transmission capabilities. The editor believes that data cross-chain transfer can achieve relatively static data verification. For example, a lending platform on the Cosmos chain wants to know your borrowing activities on the Ethereum chain to assess your creditworthiness. This can be achieved by performing simple range authentication. However, this type of data transmission has limited impact. On the other hand, dynamic data transmission may not be feasible. For example, if a lending platform on Cosmos wants to use the price on Uni as a liquidation standard, it would be difficult to achieve through Axelar’s cross-chain gateway protocol and cross-chain transmission protocol. Even if it could be achieved, it would lack timeliness. After all, transmission takes time and requires validators’ voting verification.

Note: The overall operation of the Axelar mainnet is relatively simple, and the process is clear. It primarily serves as a cross-chain transit hub for the Cosmos ecosystem and the EVM-based ecosystem. Due to differences in network programming languages and key formats, the Cosmos ecosystem and the EVM ecosystem cannot directly achieve cross-chain functionality. However, the Axelar network, built on the Cosmos SDK, can internally achieve cross-chain functionality within Cosmos using IBC. By connecting to smart contracts (gateways) in EVM-based blockchains through specific APIs, Axelar can act as an intermediary and package EVM information into the message structure required by Cosmos, enabling information transfer between the two networks [26].

5.2.2 Wormhole[27]

Wormhole is an asset cross-chain tool developed in collaboration between Solana and Certus.One, launched on September 22, 2021. As a universal messaging protocol, Wormhole can connect to multiple chains, including Ethereum, Solana, Terra, BSC, Polygon, Avalanche, Oasis, Fantom, and a total of 19 chains.

Execution Logic:

The operation logic of Wormhole is relatively simple. It is a PoS network managed by 19 validators, which deploy a Core Bridge contract on all connected networks. Wormhole Guardians run a full node for each connected chain, specifically monitoring any messages from the Core Contracts. Validators, comprising 2/3 or more, verify and sign the messages, which are then relayed to the target chain, where the messages are processed and cross-chain transactions are completed.

Unlike other bridges, the relays in Wormhole do not have special privileges. They are software that simply passes information between the Guardians network and the target chain, and are not trusted entities.

Note: It should be noted that Wormhole’s 19 validator model is relatively centralized, and only 18 validators are currently running, and the original FTX node has exited[28]. also,Wormhole has a relatively close partnership with the Jump Crypto, FTX and Solana ecosystems. Affected by the FTX thunderstorm, its future development may be affected to a certain extent.

5.2.3 Nomad[29]

Nomad is a cross-chain communication protocol that uses fraud proofs (similar to Optimistic Rollups) for cross-chain data relaying.

Execution Logic:

Nomad allows applications to send data between blockchains (including Rollups). Applications interact with the Nomad core contract to queue and send messages, which are then verified by off-chain proxies and transported between chains. To ensure the security of message delivery, Nomad uses an optimistic validation mechanism inspired by fraud-proof designs such as optimistic rollups.

Figure 5-2 Nomad execution process[30]

Nomad uses two contract addresses located on different chains (called the main contract and the replica contract) and four different off-chain participants who receive incentives to send messages across chains.

Taking the user sending a message from Ethereum to Polygon as an example, the specific simplified process is as follows:

1) The user on Ethereum submits a message to the main contract address on Ethereum. The main contract collects this message and adds it to a Merkle tree queue along with other received messages.

2) At this point, an updater, an off-chain participant, signs the message group (Merkle tree root) to update the state of the main contract. To sign these messages, the updater must stake a collateral with the main contract, which will be forfeited if any malicious behavior is proven later.

3) A relayer reads this root and forwards it to the target chain, Polygon, and then publishes it to the replica contract.

4) After the relayer publishes it, a 30-minute fraud-proof window opens. During this period, observers monitor the main contract on Ethereum and the replica contract on Polygon to ensure that all messages are correctly recorded and sent. If an observer detects malicious behavior, they can provide proof of fraud and prevent the data from being passed through.

5)If no fraud proof is submitted by the observer within the 30-minute window, the Nomad cross-chain bridge assumes that the message has been correctly recorded and sent. At this point, a processor propagates the message from the Polygon replica contract to the final recipient of the message.

Key insights: Nomad introduces a new mechanism to the cross-chain industry with an optimistic verification bridge, which allows for trade-offs between delay (or speed) and security in the design space. Overall, it provides a “lighter” user experience with weaker trust assumptions, lower costs, and more. However, the trade-off is the existence of a 30-minute delay for fraud proofs.

Due to this drawback, Nomad collaborates with a solution that provides temporary liquidity while waiting for the settlement of the cross-chain bridge - Nomad partners with Connext, which incentivizes LPs on Connext to provide short-term liquidity during the waiting period. However, LPs on Connext are exposed to the risk of malicious transactions. Additionally, Nomad has previously been hacked for $190 million, although it has since restarted, the trust in it has been compromised for the community.

5.2.4 Celer Inter-chain Message (IM)[32]

Celer Inter-chain Message (Celer IM) is designed as a “plug-and-play” cross-chain composability solution for building cross-chain dApps.

Execution Logic:

Figure 5-3 Celer IM operation process one[33]

1) User initiates a transaction to the dApp

In Celer IM, users now interact with a new dApp plugin contract (Process A in the diagram) instead of directly interacting with the existing dApp smart contract. This allows them to express their intent to execute cross-chain logic. The dApp plugin becomes part of the entire dApp business logic and may interact with existing smart contracts on the source chain. This is typically the only transaction sent by the user to interact with the cross-chain dApp.

2) dApp Plug-in sends messages and associates cross-chain transfers

After completing the necessary operations on the source chain, the dApp plugin sends the generated funds and related messages to the target chain (Process B, C in the diagram). As shown in the diagram, the Celer IM plugin contract divides the user request into two parts: token information sent to cBridge and message information sent to the Message Bus.

The message specifies the operation to be executed on the target chain. In the example of a DEX, it could be “swap token B to token C and transfer token C to the user”. By simply calling sendMessageWithTransfer, the message and fund transfer are automatically associated. Then the message is sent to the Message Bus contract, and the fund transfer is sent through the asset cross-chain bridge, which in this case is cBridge.

3) State Guardian Network (SGN) routes messages and cross-chain fund transfers

First, let’s understand what SGN is - SGN is a PoS blockchain built on Tendermint, acting as a message router between different blockchains. Node providers must stake CELR tokens to join the SGN consensus process as validators. SGN uses the same security mechanism as L1 blockchains like Cosmos and Polygon PoS chains. SGN’s CELR staking and slashing mechanisms are implemented on the Ethereum L1 smart contract.

SGN Staking nodes continuously monitor transactions occurring on all chains. Message Bus and cBridge relay information to SGN (Processes D, E in the diagram). After confirming that the message and token transfer have occurred on the target chain, SGN verifies the transaction by signature and sends it to the cBridge contract (Process F), triggering the fund transfer to the dApp plugin contract on the target chain (Process G).

Validators, on the other hand, will first reach consensus on the existence of the message and simultaneously generate a stake-weighted multi-signature proof. The proof will then be stored on the SGN chain and wait to be relayed to the target chain via an Executor that subscribes to the message (Process H).

4) Executor executes cross-chain application logic

The task of the executor is to read the equity-weighted multi-signature proof from the SGN blockchain and simply relay it to the Message Bus on the target chain (Process I). Anyone can run the executor for any application as its function is only to relay messages.

The function of the Message Bus is to check the validity of the proven messages and verify if the dApp plugin (Process J) indeed received the corresponding payment. Then, it passes the message (logical execution instruction) to the dApp Plugin contract, which hosts the cross-chain business logic of the dApp on the destination chain (Process K).

The dApp Plugin only needs to implement the executeMessageWithTransfer interface. In the DEX example, this function will execute the “swap token B to token C” on the target chain.

Furthermore, Celer IM does not necessarily use fund transfers to send cross-chain messages or logical execution instructions. For example, in an NFT marketplace, if a user participates in an auction happening on different chains, they only need to lock their funds without actually transferring the assets to the target chain for bidding. Fund transfer is only required if they win the auction. The process is as shown below:

Figure 5-4 Celer IM operation process 2

Note: The above process is excerpted from the officially released “Celer Inter-chain Message Framework: the Paradigm Shift for Building and Using Multi-blockchain dApps”. Some of the content has been deleted. For details, please refer to the original text (requires scientific Internet access).

Perspective: After SGN as a public liquidity pool for cBridge 2.0 (2022.03), users who do not operate nodes can also provide liquidity for cBridge, making it more convenient for Layer2 or other Layer1 projects to provide liquidity on Celer, which is beneficial to increase the liquidity depth of cBridge. SGN, as a node gateway and arbitrator, also helps Bridge provide better services. Looking at the dashboard of cBridge 2.0, its TVL did experience a rapid growth in March-April 2022, but with the LUNA incident in May and subsequent market downturn, the current TVL has dropped to the range of $150-200 million.

Overall, Celer IM’s security assumptions are built on its PoS chain and have two security models: optimistic-rollup inspired (not mentioned above, interested readers can refer to it themselves) and L1-PoS-blockchain security, which users and developers can freely choose and set. It performs well in terms of security. In addition, although the cBridge economic model has seen good improvements compared to v1, it is also due to its PoS mechanism that Celer IM heavily relies on CELR through staking. Users of Celer IM must pay CELR fees to SGN for cross-chain consensus services. If the price of CELR tokens significantly declines, the security of SGN is likely to also decrease [34].

5.2.5 anyCall of Multichain[35]

anyCall is a universal cross-chain messaging infrastructure for exchanging arbitrary data. It consists of a smart contract system and Multichain’s SMPC network, which is a secure multi-party computation validator network.

Execution Logic:

In anyCall, the validator network can access contracts on different chains and verify the transmitted information between these contracts. It completes the reception and transmission of information, sending any transmitted information to the target chain specified by the business logic and triggering subsequent smart contracts to implement the business logic. The specific process is as follows:

1) The dApp needs to deploy a sender contract on Chain A (source chain) and a receiver contract on Chain B (target chain). On the receiver contract, there needs to be an anyExecute function that will be called.

2) When the dApp sends a message by calling the sender contract, the anyCall contract verifies the message and relays it to the target chain.

3) Multichain’s MPC network (consisting of 24 nodes) is responsible for validating the messages sent to the anyCall contract by the anyCall function. The anyCall contract exists in the public MPC address of all supported blockchains. When the anyCall function sends a message, MPC nodes ensure the security of the message before sending it to the target chain.

4) After successful verification, the anyExec function receives the message from the anyCall contract and executes the request on the target chain.

Key Takeaways: The trust assumption of anyCall relies heavily on Multichain’s MPC network, so users need to trust that the nodes will not act maliciously. Mechanically, compared to similar AMBs, it can be considered relatively simple and more centralized. However, Multichain’s scale has consistently been at the forefront of all cross-chain bridge races. It should be noted that Anyswap has suffered hacking attacks during the iteration from Anyswap to Multichain.

5.3 Competitive analysis

Above, we have listed five types of Arbitrary Message Bridge (AMBs), and it can be seen that each type of cross-chain bridge has its own trade-offs.

Axelar, Wormhole, and Multichain’s anyCall all use external validation methods to facilitate the transmission of arbitrary information between any two public chains through their own PoS chains/networks. The advantages are fast speed, low fees, and the ability to interact with the data on any number of target chains, making it easier to connect to more chains. However, the downside is that this approach sacrifices security and requires users/LPs to fully trust the external validators’ funds/data, relying on the bridge’s security rather than the source or target chain.

There are differences in the specific split. For example, in terms of validator permission, Axelar only allows 50 validators as the sole active set on the entire network. To become a formal validator, tokens must be ranked within the top 50. However, any user can delegate their tokens to the corresponding node. In anyCall, anyone can run their own MPC node. In Wormhole, only Guardians with permission can become validators.

The Celer IM architecture is supported by a combination of on-chain smart contracts for receiving and sending messages and the Celer PoS network. Although the security assumption is also based on its PoS chain, Celer IM has two security models: optimistic-rollup inspired (where malicious cross-chain messages are not processed as long as there is one application supervisor that remains honest and operates normally) and L1-PoS-blockchain security. Users and developers can freely choose and set these models.

Nomad uses fraud proofs (similar to Optimistic Rollups) for cross-chain data relays, introducing new trade-offs in the field of cross-chain bridges, exchanging delay (or speed) for security.

In addition, users with different fund sizes have varying considerations for fund efficiency and security systems. Each bridge focuses on a specific area and has corresponding user needs. Overall, the current Arbitrary Messaging Bridges (AMBs) are still in a very early stage, making it difficult to directly compare these “bridges” in terms of superiority. It can only be said that they each have their own advantages and disadvantages in different dimensions.

For a detailed comparison of the aforementioned AMBs, you can refer to the article “Navigating Arbitrary Messaging Bridges: A Comparison Framework” [36] written by Arjun Chand, a member of the LI.FI team. The article provides a comprehensive comparison of the above projects from multiple dimensions, so this article will not provide further description.

• LayerZero

Comparing to the Arbitrary Message Bridge (AMB) described above, LayerZero has a major difference in that it does not require running nodes on connected chains, outsourcing the burden of verifying on-chain information transfer to third parties such as oracles. This approach makes the protocol lighter and reduces operational costs in the early stages. We can now see that LayerZero is rapidly expanding in the early stages of the project, leveraging its own advantages.

With the emergence of LayerZero, it opens up another path for us, not just continuously optimizing the performance of bridges, but abstracting the chains from users.

Specifically, previously if we wanted to transfer assets between two different chains, we needed to go to the UI of a third-party cross-chain bridge and transfer our assets to the target chain. However, in many cases, cross-chain bridges do not support the cross-chain transfer of our altcoins, so we often need to perform several additional swaps to successfully migrate assets to the target chain, which can be cumbersome in terms of operations.

Based on Stargate built on LayerZero, its core is to let current DApps (such as Uniswap, Sushi, and other DEXs) integrate cross-chain bridge protocols, allowing users to directly schedule and cross-chain assets through the DApps they are currently using.

For example, SushiSwap is deployed on 18 chains, and it is difficult to share a global state. If we use the previous solution, we would need to deploy a bridge between each pair of chains. However, by utilizing the LayerZero protocol, we only need to use the endpoint of each chain to share the global state[37].

For instance, when SushiSwap integrates Stargate, in this case, if a user wants to exchange wBTC on Ethereum for MATIC on Polygon, the user can execute this operation in a single transaction on the source chain without leaving the SushiSwap UI. This provides standardized experience for multi-chain DApps such as SushiSwap and Uniswap. In the author’s view, this is an ideal cross-chain method that greatly improves the usability of asset cross-chain transfers.

So is LayerZero’s solution better than other AMBs? Not necessarily. The security of the LayerZero protocol still needs to be validated by the market. And bridges like Axelar and Celer IM, which build bridges from scratch, although they have high costs and long cycles, to some extent, they also have a more sustainable foundation for expansion and greater value accumulation. If Nomad had not been subjected to hacker attacks, its unique improvements based on fraud proofs might have been widely adopted by the market.？

In summary:

Looking at the development trend of cross-chain bridge projects in the past two years, we can see a clear main theme, which is that most of these projects are continuously developing around the goal of building a more “robust bridge”. Ultimately, this is about how to better achieve the three elements: security, seamlessness, and speed. The race is still evolving, and the future of who will become the preferred solution for multi-chain has just begun.

In conclusion, although LayerZero has a strong narrative, there are still many details that have not been fully disclosed, and there are corresponding risks (see the product section for details). Additionally, LayerZero achieves cross-chain concepts through oracles and relay transmission of information, a concept that has already been addressed by Chainlink’s Cross-Chain Interoperability Protocol (CCIP). According to existing information, Chainlink may become a favorable competitor to LayerZero. However, the CCIP concept has been silent for a long time since its release, with no whitepaper published yet, and their developers seem to be continuously working on its development. A comprehensive comparison between Chainlink CCIP and LayerZero has already been made by Pickle and Aylo (pseudonyms), so this article will not provide further description. For more details, you can refer to their article

6.Risk

Protocol security

The security of LayerZero has not been fully verified. The trust assumptions that the oracle and relayers need to operate independently from each other are questionable. The security risks behind the relaying mechanism still need to be monitored. For more details, please refer to section 2.4.2 Security in the aforementioned product.

Unknown Tokenomic Model

The economic model of LayerZero has not been released yet. It still needs further observation in the future.

Disclaimer：

1. This article is reprinted from [头等仓区块链研究院]. All copyrights belong to the original author [头等仓]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. Translations of the article into other languages are done by the Gate Learn team. Unless mentioned, copying, distributing, or plagiarizing the translated articles is prohibited.