Summary
LlamaRisk preliminarily supports onboarding MegaETH, an Ethereum L2, to Aave while underscoring that its ecosystem is still in the testnet phase (launched on March 6, 2025) and lacks key infrastructure components necessary for Aave such as audited codebase, Chainlink oracles, and a bug bounty program. However, according to the team, some are expected to roll out soon. MegaETH aims to achieve 100,000 TPS, 10 Ggas/sec, and sub-10 millisecond block times, using node specialization, state diff propagation instead of syncing entire state changes, and a stateless validation model.
However, these come with trade-offs, including reduced decentralization (a single active sequencer at a time ordering transactions) and security risks from untested tech at scale like stateless validation and EigenDA (external component built on EigenLayer). As the infrastructure matures and the mainnet launch approaches (tentatively projected for the end of Q2 to the start of Q3, per the team), a more precise assessment of the project’s ecosystem and real-time performance will be possible. LlamaRisk will provide asset recommendations once the network is closer to mainnet readiness.
1. Network Fundamental Characteristics
1.1 Network Overview
MegaETH is building on OP Stack customized to work with Reth. It qualifies as an optimistic validium, relying on Ethereum for settlement and EigenDA for data availability. MegaETH adopts node specialization, allowing specialized operators like sequencers to handle the heavy lifting. It uses a centralized single active sequencer because it is needed to ramp up throughput and optimizations so that the L2 can scale up to 100,000 TPS and 10 Ggas/sec.
MegaETH has two types of blocks: mini blocks and EVM blocks with 10ms and 1s block times, respectively. EVM blocks ensure backward compatibility with existing tools and applications, while Mini blocks are MegaETH specific and require its Realtime API to access the information. Both block types share the same rollback guarantees, but finality depends on factors like Ethereum reorgs, DA failures, sequencer downtimes, etc. For MegaETH, implementing robust mechanisms to mitigate these risks is crucial to maintaining the rollback guarantee for 10ms Mini Blocks.
Source: MegaETH Major Components, MegaETH.
List of modifications from MegaETH that help them achieve higher throughput and lower block times:
- Unlike Ethereum and other L2s, where full nodes perform both consensus and execution, MegaETH introduces node specialization (four specialized node types), which eliminates consensus overhead in execution and reduces redundant computation, allowing real-time updates.
- Sequencer: An HPC machine responsible for ingesting and ordering transactions.
- Replica Nodes: Receive state diffs via a P2P network, enabling rapid state updates without reprocessing transactions.
- Prover: Implements the stateless validation scheme to validate blocks asynchronously and out-of-order.
- Full Node: Re-executes all transactions to validate blocks.
- Replica Nodes apply state diffs instead of every node executing transactions, leading to faster state updates and improved efficiency in state synchronization.
- MegaETH uses a centralized sequencer like some L2s but goes further by having only one active sequencer at a time. This significantly improves transaction throughput by eliminating network-wide consensus bottlenecks. However, MegaETH’s sequencer node hardware requirement is high (100-core CPUs, 1-4TB RAM, AWS r6a.48xlarge, $10/hr).
- Full nodes validate transactions asynchronously using stateless validation proofs instead of synchronously re-executing each transaction, reducing computational overhead on the network.
- EigenDA is used for data availability instead of Ethereum’s native solution to reduce costs and data storage efficiency.
However, some trade-offs and risks arise from these modifications:
- A single active sequencer centralizes transaction ordering, making megaETH vulnerable to outages and MEV manipulation or censorship.
- Replica nodes applying state diffs without re-executing transactions can lead to incorrect state propagation if proofs are compromised.
- The stateless validation model is relatively new and untested at scale, posing risks of unforeseen issues or exploits. Similarly, reliance on EigenDA for data availability is a new solution.
- Ethereum’s trustlessness (full nodes executing all transactions) is sacrificed by having nodes accept state diffs rather than recomputes, introducing trust assumptions in the proving mechanism.
Previously, MegaETH’s launch-pool contract (currently unavailable) was audited by OtterSec on August 16, 2024. The audit identified four findings: two critical and two informational, all of which were resolved. Though they’re building on OP Stack and Reth, given the scale of their customization, it would be essential to see publicly available audits once the code is made available as an open source.
1.2 Decentralization and Legal Evaluation
MegaETH’s decentralization remains uncertain as key infrastructure, including testnet development and security mechanisms, is still evolving. While it has been confirmed that the sequencer will remain centralized to reduce computational overhead, they plan to implement a rotating system with sequencers distributed across multiple geographic locations, each with several hot backups. The exact number of sequencers has not been finalized, as the network is still in the testnet phase. Additionally, details on governance participation, including decision-making on key proposals and network upgrades, remain unclear at this stage.
Legal Commentary
The team reported that a Cayman Islands foundation is being established to supervise MegaETH governance. At this juncture, no official terms of use or other legal documents are available that clarify the foundation’s (or any other legal entity’s) degree of control over the sequencer.
2. Network Market Outlook
2.1 Market Infrastructure
MegaETH is currently in Testnet and focuses on onboarding developers and infrastructure providers. In the meantime, the public faucet is disabled. The team announced their deployment timeline: testnet deployment on March 6, 2025, app and infra onboarding from March 6-10, and user onboarding starting March 10. Builders will initially access 1.68 Ggas/sec with 15ms effective block times on a single-threaded setup, with further scaling planned.
Source: MegaETH block time since launch, MegaETH, March 11, 2025.
The chart shows multiple spikes in block time exceeding the baseline 10ms. Following the launch, MegaETH’s RPC node experienced downtime due to an edge case vulnerability, indicating that thorough testing will be required before further upgrades are launched.
DeFi applications building on MegaETH:
Tooling
Two block explorers are available: the official real-time performance dashboard uptime.megaeth.com and the community-run megaexplorer.xyz.
MegaETH has built in-house RPC nodes to support its high-speed environment, available at carrot.megaeth.com/rpc. However, the RPC has experienced downtimes post-launch, indicating the need for further optimization.
At this stage, details about the indexer and Oracle infrastructure remain unavailable, suggesting that these components are still in development or yet to be integrated.
2.2 Ecosystem Growth Potential
On February 12, 2025, MegaETH raised $13M through its NFT mint, “The Fluffle.” Earlier, on December 13, 2024, it secured $10M via the Echo sale, attracting over 3,200 investors from 94 countries. Previously, on June 27, 2024, MegaETH raised $20M in a Seed round from notable investors, including Vitalik Buterin and Dragonfly.
The team has been working on the World Computer thesis, and to evolve it, they’ve been closely working with builders through their accelerator program “MegaMafia,” which has been a success as the projects building on it have raised more money than the chain itself. This highlights the ecosystem’s growing infrastructure and future potential.
Source: MegaETH Testnet Roadmap, MegaETH, March 11, 2025.
The roadmap indicates that the testnet infrastructure is still in development and will require more time to mature, suggesting that the mainnet launch is not imminent. Past L2s have typically taken 4 to 10 months from testnet to mainnet, depending on testing requirements and audits, for example, Optimism (10 months), Linea (4 months), and Base (6 months). The MegaETH team has stated that the mainnet launch is expected between the end of Q2 and the start of Q3.
3. Impact of Aave Deployment
MegaETH is EVM-compatible, allowing Aave to deploy using the same smart contracts. However, optimizations would be necessary to accommodate its high-speed environment, mainly if mini blocks are used, as they have significantly lower block times (10ms) than EVM blocks.
List of lending markets building on MegaETH:
- Teko: Real-time lending market with granular micro-liquidations, margin trading, and under-collateralized borrowing, all with the same deep onchain liquidity.
- Avon: Orderbook matching-based lending framework with lending pools at the strategy layer, demand-based rates, and sub-second liquidation.
The common thing noticed here is MegaETH’s high throughput, enabling frequent Oracle updates and quicker liquidations, allowing for a more capital-efficient deployment. However, the description is theoretical, with no real test on the security and efficiency of such systems in production.
Currently, Chainlink oracles, the preferred solution for price feeds, are not integrated.
Disclaimer
This review was independently prepared by LlamaRisk, a community-led decentralized organization funded in part by the Aave DAO. LlamaRisk is not directly affiliated with the protocol(s) reviewed in this assessment and did not receive any compensation from the protocol(s) or their affiliated entities for this work.
The information provided should not be construed as legal, financial, tax, or professional advice.
