What are the real use cases that the first batch of AVS can provide?
Restaking ecosystem has finally reached two important milestones: the launch of EigenLayer (and EigenDA) on the mainnet, and the first batch of AVS (AltLayer, Brevis, eoracle, Lagrange, Witness Chain, Xterio) going live on the mainnet.
AVS is the ultimate embodiment of the practicality and security of the EigenLayer protocol. The previous Restaking protocols (Renzo/Puffer, etc.) or liquidity re-staking tokens were just the starting point and a means to attract liquidity for this ecosystem. Previously, more ETH entered the Restaking ecosystem and more nodes were mobilized through incentive supply. AVS has the potential to release demand and provide more value to crypto and decentralized networks.
The first batch of AVS went live in various tracks, including Rollup-as-a-Service, oracles, ZK protocol processing, DePIN, and gaming. The most noteworthy is the ZK co-processor, which is a relatively new concept without mature products. EigenLayer surprisingly supports both Brevis and Lagrange in the first batch.
Furthermore, the launch of EigenLayer’s mainnet does not mean that the protocol has entered a mature stage. There are still many modules and solutions that are not fully understood, such as the slashing mechanism for future nodes, ensuring the security of AVS from an economic perspective, and even the design of token economics, which need to be disclosed by the team in the future.
AVS, which stands for Actively Validated Services, is a concept set in the EigenLayer protocol. In simple terms, AVS can be analogized as “middleware” that provides services to end products, such as data and verification capabilities mentioned frequently as “oracles” which are not end products themselves but can provide data services to DeFi, gaming, and wallets, which is one type of AVS.
The downstream of AVS is likely to be an end product that directly faces ordinary users. The upstream of AVS consists of nodes participating in Restaking, who support specific AVS by collecting ETH in protocols like Puffer/Renzo. Compared to most protocols, EigenLayer’s business model is relatively direct. End users will directly or indirectly pay for the products they use, and these fees will be allocated to AVS, node operators, the EigenLayer protocol, and users providing Restaking ETH. The specific distribution may vary, and in the early stages, the advantages of crypto “tokenomics” can be utilized to reward users with the protocol’s own tokens.
Therefore, it is necessary to develop more types of AVS in the future and ensure the reliability of AVS services trusted by end products in order to complete the entire ecosystem.
AltLayer is a Rollup-as-a-Service (RaaS) provider that can customize the deployment of Layer2 networks such as Rollups according to demand. For a Rollup network, the choice of DA is crucial, so AltLayer supports EigenDA developed by EigenLayer in addition to Ethereum.
In addition, AltLayer has collaborated with EigenLayer to release the Restaked Rollup framework and provides three modular AVS: VITAL (AVS for decentralized verification of Rollup’s state), MACH (AVS for fast finality), and SQUAD (AVS for decentralized sequencing). These AVS aim to solve the problems of slow finality, centralization in settlement layers and sequencers on the blockchain. The MACH module, which provides services for Xterio and Optimism, was launched this time.
Brevis provides a solution that enables the capability of a ZK co-processor through AVS. Its team includes members from the cross-chain bridge protocol Celer Network, and Mo Dong is the co-founder of both projects. This time, he also gave a presentation on “A Smart ZK Coprocessor” at the Web3 Scholars Conference in Hong Kong.
Brevis proposes the coChain solution to further reduce the cost of implementing a “ZK co-processor” based on smart contracts and zero-knowledge proof technology. It provides a more cost-effective solution and enables capabilities that cannot be achieved with a fully smart contract and zero-knowledge proof-based approach. After all, EVM has many limitations and constraints. Brevis coChain is a PoS blockchain that can protect its security through ETH staking, relying on the EigenLayer protocol. It is more like a combination of optimistic mechanism and ZK mechanism, or can be called fraud proof and validity proof. If any malfeasance is detected, a challenge can be initiated by generating zero-knowledge proofs and punishing the malicious party. This also involves some game theory and considerations in token economics design.
eoracle, the oracle protocol, derives its name from (e)thereum + oracle. They consider themselves as the first “native” oracle on Ethereum, possibly because the security of this oracle is guaranteed by staked ETH, while other oracles like Chainlink rely on Chainlink’s node network and its token LINK for security assumptions.
The demand and business model of oracles are much clearer compared to other AVS. Many DeFi and RWA require off-chain data, and oracle networks verify data through participating nodes.
eoracle explicitly states that it will adopt a dual-token model. Other AVS may also adopt this model, which means the security of the network relies on Restaked ETH and native tokens of AVS are issued to incentivize nodes. The specific uses and design of native tokens have not been disclosed yet, but they believe native tokens can enhance network participation (incentivize users?), ensure fair value distribution (income distribution based on tokens?), and promote the decentralization of the eoracle protocol (as weight or governance?).
Lagrange is also a ZK co-processor, but they emphasize the concept of “parallel”. It is somewhat similar to the services provided by Brevis.
Lagrange’s team states that their designed ZK co-processor natively supports parallelization and horizontal scaling, enabling easy proof of large-scale distributed computing results stored on the chain or transaction data. The proof workload can be distributed among thousands of working nodes simultaneously, and security is guaranteed by ETH on EigenLayer.
Last month, Renzo, Swell, and Puffer also announced their collaboration with Lagrange, with each commissioning $500 million worth of Restaked ETH to Lagrange. Lagrange has also designed some features for these platforms based on their protocol characteristics, such as the ability to call Lagrange to obtain on-chain historical data and calculate scores for users based on this data.
Witness Chain is a network designed specifically for decentralized IoT devices. It includes various components, such as DCL (DePIN Coordination Layer), which provides basic services needed for the DePIN ecosystem, such as the security of the chain itself, node bandwidth, and physical location. They call these basic services “watchtowers” used to measure the aforementioned data and generate valid proofs and utilization in the DCL layer. This is similar to the literal meaning of “witness” in Witness Chain: “to testify”.
Xterio is slightly different from other AVS mentioned above. It is actually a Layer2 blockchain based on EigenDA and OP Stack issued through AltLayer’s RaaS. Xterio Chain will focus on AI and Web3 gaming-related scenarios. AltLayer states that Xterio L2 uses the aforementioned MACH (AVS for fast finality). AltLayer also provides MACH services to the mainnet of Optimism.
EigenLayer will definitely have more types of AVS going live. However, EigenLayer also raises systemic risks to the Ethereum ecosystem, which concerns many people because EigenLayer bypasses “smart contracts” and directly takes over the node ecosystem of Ethereum, which is different from all previous Ethereum-based protocols. However, this is also the charm of permissionless systems. Even without EigenLayer, others would attempt this direction.
In addition, Lido, as the largest liquidity staking protocol in the Ethereum ecosystem, not only stakes the most ETH but also has many node operators. The conflicting interests between EigenLayer and Lido may make Lido reconsider its business model and sustainability, and EigenLayer itself needs time to gradually fill in the missing modules.
AltLayer
AVS
Brevis
ChainFeeds
EigenLayer
eoracle
Lagrange
Witness Chain
Xterio
