Author: YBB Capital Researcher Ac-Core
Metis is an Ethereum L2 ecosystem based on Optimistic Rollup. It launched the Andromeda mainnet early. Compared with other L2 networks, its biggest feature is the decentralized sorter, which makes the transaction sorting rights belong to community participants. In March 2025, Metis officially announced the dual-chain strategy: based on the existing general chain Andromeda, it launched a high-performance Hyperion chain. The former is positioned as a safe and reliable general L2, while the latter focuses on high-frequency, high-throughput and AI-driven application scenarios. Hyperion is built on the Metis SDK, and its core is the new MetisVM virtual machine (compatible with EVM to support AI optimization instructions). The goal is to optimize scalability and decentralization, while greatly improving transaction efficiency.
According to Metis officials, Hyperion is committed to building a "high-performance, AI-native" L2 network that can achieve near-millisecond transaction confirmation and ultra-high transaction throughput, making AI-native applications (such as on-chain LLM reasoning and decentralized AI agents) possible. In general, Metis uses a dual-chain architecture, with Andromeda emphasizing security and decentralization, and Hyperion emphasizing high performance and AI applications, and the two provide complementary support for different application scenarios.
Overall, Metis's development path can be summarized into two points: First, continue to improve the core L2 infrastructure, such as adding real-time data availability migration and fraud proof mechanism in the Andromeda upgrade in May 2025, making it the industry's first "truly decentralized" Layer 2; second, move towards modularization and multi-chain interoperability, lower the development threshold through Metis SDK, and promote the development of multi-chain ecology. The launch of Hyperion has reshaped the value logic of the METIS token to a certain extent: making Metis no longer just an L2, but transforming into a "multi-chain infrastructure + AI dedicated chain" platform.
Image source: Metis
As an Optimistic Rollup, Hyperion inherits the security model of Metis and Optimism, and mainly changes in parallel computing, data availability and decentralization. The main optimization direction is AI and high-frequency application scenarios. Specifically, Metis promotes ecological construction through the three major systems of Metis SDK + Hyperion + LazAI.
Parallel execution engine: Traditional blockchains are usually executed in order of transactions. Hyperion introduces parallel execution technologies such as Block-STM to allow independent transactions in a block to run together. The use of optimistic concurrency control and dynamic DAG scheduling algorithms allows independent transactions in the same block to run simultaneously, greatly improving throughput;
Decentralized sorter: Hyperion's transaction sorting rights are shared by a multi-node network, rather than controlled by a single node or a few operators. Specific mechanisms include Leader Rotation and Encrypted Mempool combined with Proposer-Builder Separation (PBS) to prevent front-end transactions and centralized intervention;
Data availability: Hyperion plans to make full use of Ethereum's new features and external DA services to ensure data availability. As Ethereum EIP-4844 progresses, Hyperion (and Andromeda) will publish transaction data directly to L1 through blob transactions to inherit Ethereum's security and reduce costs. According to official statements, Metis is also preparing to integrate EigenDA;
Fraud proof mechanism: Hyperion continues to use the optimistic model, but has upgraded the fraud proof. Metis promised to introduce a complete fraud proof mechanism and data availability solution early on; its ReGenesis roadmap also proposed "implementing the latest fault proof system from OP Stack." In other words, Hyperion will adopt an interactive error correction mechanism: when the sorter publishes a new block, the "observer" in the network has a window period to submit a challenge (such as a 7-day period) to prove that the block is invalid. Once this challenge is successful, the malicious block will be rolled back and the responsible party will be punished.
AI native infrastructure: Metis SDK allows users to build L2 or L3 blockchains by themselves, which has integrated an upgraded version of EVM (called MetisVM) and parallel processing capabilities, and is also equipped with a decentralized sorting mechanism. Hyperion uses this SDK to build a high-performance AI-specific Layer2, which combines Optimistic Rollup, parallel execution and distributed sorting, and can achieve almost real-time transaction confirmation and on-chain AI reasoning. According to official statements, MetisVM has improved transaction efficiency by about 30% through dynamic optimization of opcodes and parallel execution mechanisms. At the same time, MetisDB uses memory-mapped Merkle trees and concurrency control, and state access can be done at the nanosecond level, almost eliminating the storage bottleneck. These technologies combined allow Hyperion to run AI reasoning tasks such as large language models (LLM) directly on the chain, laying a strong foundation for AI contracts.
AI narrative project incubation: Metis is also actively incubating AI-related projects, such as the LazAI protocol. This is an open network focused on "trusted AI data assets" with the goal of solving the problem of opaque and inconsistent data used by AI. LazAI uses the verifiable computing and standardization mechanisms of blockchain to create an open and transparent data market to ensure that AI models use high-quality data and support cross-chain use. Based on this protocol, Metis also launched the Alith framework, an AI agent development tool tailored for blockchain. Developers can use the Alith SDK to quickly write AI Agents on Metis and deploy them online. The official also gave a very down-to-earth example: users only need to send instructions in natural language through the Telegram chatbot to complete DeFi operations such as lending, without having to write smart contracts themselves. This design greatly reduces the threshold for ordinary people to use AI applications and makes the work of developers much easier.
Hyperion is still based on Optimistic Rollup. Metis has built differentiated competitiveness through forward-looking technology and strategic layout: Hyperion and other technologies meet the performance requirements of AI applications, dual-chain and SDK models take into account both general and professional needs, and decentralized sorting improves ecological trust. The differentiated competition with other L2 is mainly reflected in the Hyperion architecture, dual-chain strategy, Metis SDK and decentralized sorter:
High-performance execution layer:Metis Hyperion is a high-performance Layer2 for AI. Officials say that it has achieved near-real-time settlement and Web2-level response speed through Optimistic Rollup, parallel execution and distributed sorting technology. Hyperion introduced MetisVM, a virtual machine customized for high-frequency trading and AI tasks, with dynamic opcode optimization, parallel execution and caching mechanisms, which can greatly improve the efficiency of smart contract execution. At the same time, Hyperion designed on-chain reasoning support for AI reasoning, including special pre-compiled contracts and execution engine optimization;
Dual chain strategy (Andromeda + Hyperion):Metis retains the original Andromeda chain as a general-purpose L2, and continues to provide stable infrastructure for applications such as DeFi; at the same time, Hyperion is launched to serve AI scenarios. This "dual network" architecture solves the trade-off between versatility and professionalism: it avoids stalemate with other L2s in homogeneous competition, and also finds exclusive carrying space for AI applications;
Metis SDK and developer ecology:Metis SDK is a modular toolkit for developers that integrates drawings, construction tools and standard interfaces, allowing developers to quickly build custom execution layers or applications. Metis SDK allows developers to quickly deploy Layer2/Layer3 with the help of mature modules, and also facilitates smooth migration or interaction between Andromeda and Hyperion, which effectively improves the efficiency and interoperability of ecological construction;
Decentralized sorter: In 2024, Metis successfully launched the industry's first fully decentralized sorter, which handed over the transaction sorting rights to the community nodes and staking mechanism. The mechanism achieves fault tolerance and anti-censorship capabilities through rotating consensus nodes and incentive token governance, completely eliminating the risk of single point failure.
Image source: investx.fr
Assuming that Ethereum will focus on the development of its own L1 (such as underlying consensus and sharding upgrades) instead of continuing to directly expand L2, how can L2 platforms like Metis maintain and expand their influence? In the All in AI strategic roadmap released by Metis, at this stage, Metis's strategy is to take a modular and multi-chain route.
First of all, Metis's dual-chain architecture and MetisSDK provide it with the ability to build multiple dedicated chains. At the same time, the launch of Hyperion marks Metis' transformation from a "single L2" to a "modular multi-chain infrastructure." Through MetisSDK, any team can quickly create a customized blockchain like building blocks - configuring parallel execution consensus, EVM compatibility layer, AI optimized VM, on-chain storage and other components. This means that Metis not only operates two chains, but also supports more "industry-specific chains": such as AI computing chain, DePIN infrastructure chain, and game chain, each meeting the needs of different scenarios.
Secondly, Metis attaches great importance to cross-chain interoperability and collaborative ecology. The official roadmap and community communication have repeatedly mentioned the introduction of cross-chain bridging and data/computing aggregation mechanisms. It is not difficult to find that Metis will integrate Chainlink CCIP, so that assets and smart contracts can flow freely between Metis and other public chains. The Hyperion architecture also emphasizes "shared bridge + cross-chain interconnection" and proposes "decentralized data and computing aggregation" to connect AI applications with various data networks and computing resources.
The community is not only a user, but also a direct participant and beneficiary of network operation. By opening the role of the sorter, anyone can become a block proposer through staking and obtain sorting rewards. At the same time, Metis is planning to introduce an incentive mechanism for AI node operation to encourage more developers to deploy reasoning services to provide on-chain AI capabilities.
Let's look at the problem directly: Currently, using AI services on the chain, such as transmitting results through oracles, has obvious centralization risks and performance bottlenecks. In order to solve these problems, Hyperion and LazAI have cooperated to introduce an AI agent framework called Alith. Developers can use Alith SDK to write "AI agents" and deploy them as contract modules on Hyperion. These agents can handle functions such as model selection, reasoning logic, and fault response. Other contracts on the chain can be called directly, for example, to implement chatbots, prediction tools, or DAO assistants. Through mechanisms such as pre-compiled contracts, Hyperion incorporates AI reasoning into the on-chain execution process, and the reasoning results can also be fixed on the chain through logs, reproducible operations, or trusted execution environments, so that they can be verified and trusted. This not only preserves the transparency of the blockchain, but also meets the computing needs of AI, and truly promotes "AI on the chain."
In addition, Hyperion's parallel processing and low latency are particularly suitable for AI tasks. As long as there is no conflict between requests, they can run simultaneously, and the efficiency is much higher than the traditional L2 sequential execution. MetisVM has also made special AI optimizations, such as using Rust/WASM to improve performance, supporting multiple inputs such as text and images, so that AI models can run directly on the chain. This entire set of designs is to achieve what Metis calls "the first Layer2 protocol that can run large language models locally on the chain."
Image source: self-made
The market enthusiasm for Crypto+AI has undoubtedly achieved great success on Solana. The Solana community is also promoting open protocols such as Model Context Protocol (MCP), trying to enable off-chain AI models to query on-chain data through standardized interfaces. QuickNode demonstrated how to build Solana's MCP server, enabling "orthodox" AIs such as Claude to query on-chain data through RPC Directly access the Solana blockchain information. However, most AI solutions on Solana run AI models off-chain and call the results on-chain.
As pointed out in "Solana and Base, which ecosystem is more suitable for AI Agent?": "Currently, all models including agents are running off-chain, and input data, training, and output information are not on-chain; EVM chain or Solana/BASE do not support the combination of AI and contracts." Solutions such as Solana+MCP do not really "put AI models on the chain to run", but only allow AI models to safely access on-chain data.
However, the interesting thing about Hyperion is that it tries to make AI reasoning itself execute on-chain, which Solana has not achieved. If Hyperion implements the first Layer-2 protocol that supports local execution of LLM on the chain, it means that Hyperion not only provides a data interface, but also directly provides on-chain computing power for AI models to run in MetisVM, and each step of the calculation is completed in the execution environment of the blockchain. Such a design decentralizes AI more thoroughly than Solana's solution. The AI reasoning results on Hyperion can be verified and traced on the chain, which is naturally resistant to tampering and review; while Solana's MPC is more about trusting external models and only ensuring the security of data paths.
Of course, Solana has a natural advantage in basic performance in terms of high-concurrency processing capabilities and mature GPU chip support architecture. Hyperion focuses on compatibility and ecological linkage with the Ethereum ecosystem: EVM compatibility, MetisSDK ecosystem and METIS token liquidity.
Overall, Hyperion and Solana are not direct substitutes, but provide another path to blockchain + AI: Solana relies on its network performance + traditional LLM interface (MCP); Metis relies on its smart contract platform as the foundation to launch native on-chain reasoning functions.
To put it bluntly, Hyperion is not the current master key. At present, most of the claimed "AI+chain" projects are still at the conceptual level, and there are very few cases that can be verified as productivity applications. Model ownership and trust issues have not yet been resolved: How can models trained outside the chain trace their sources and how can the correctness of their execution results be proved on the chain? These are all questions that need to be answered at the infrastructure level.
Hyperion's system design targets the above pain points to a certain extent. It solves some computing power problems by providing support for AI reasoning at the protocol layer: parallel execution and MetisVM optimization make the on-chain computing power far exceed the traditional L2, but this does not mean that Hyperion solves all problems at once.
However, it is certain that Hyperion has brought new possibilities for Web3 AI and provided the Ethereum camp with a bargaining chip to participate in the AI narrative. It mainly solves problems in computing architecture and trust foundation: a large amount of parallel computing power, verifiable execution process on the chain, modular tool chain, and native support for special AI needs. These efforts create the possibility for the emergence of truly practical blockchain + AI applications (such as on-chain autonomous agents and high-frequency data analysis) in the future, adding substantial content to the AI narrative in the crypto field.
El Salvador is increasing its Bitcoin holdings despite IMF conditions, recently purchasing 11 BTC and bringing its total to over 6,067 BTC. To comply with its $1.4 billion IMF agreement, the country has made Bitcoin adoption optional and removed tax payments in BTC.
WeatherlyAfter speaking with President Trump, Canada and Mexico agreed to curb illegal drugs and immigration, but negotiations continue.
KikyoNicolas Cage has expressed concerns about AI replacing human creativity in filmmaking, warning against its use to manipulate performances. While some, like Tom Hanks, see AI as a way to extend actors' careers, the debate continues to grow in Hollywood, with critical opinions on AI's role in altering films.
JoyMeta’s Frontier AI Framework restricts high-risk AI until safeguards are in place and halts critical-risk AI development indefinitely, prioritising risk mitigation before wider access.
CatherineMicroStrategy paused its Bitcoin purchases after 12 weeks due to market volatility, tax concerns, and economic uncertainty. The company still holds 471,107 BTC but is reassessing its strategy amid fluctuating prices and changing global conditions.
WeatherlyBybit’s CEO, Ben Zhou, warned that crypto liquidations may be as high as $10 billion, far exceeding the $2 billion reported. He explained that limitations in data sharing caused discrepancies, with actual losses likely much higher than public figures suggest.
AnaisLed by Commissioner Hester Peirce, the new crypto task force seeks to resolve regulatory ambiguity, focusing on defining digital assets as securities and enhancing transparency. With the SEC’s stance shifting post-Gensler, can this initiative truly end years of uncertainty?
CatherineCarnegie Mellon and NVIDIA have developed a training technique that gives humanoid robots unmatched agility, mimicking elite athletes like Ronaldo and Kobe. As AI pushes physical limits, could it outpace human athletes?
KikyoMarvel Studios faced criticism from fans who accused them of using AI-generated artwork for a "Fantastic Four" poster due to distorted hands and other anomalies. Marvel denied the allegations, stating no AI was involved in creating the poster.
AnaisThe platform will provide BTC and AVA rewards for users who book accommodations on Trivago using crypto. Could this incentive drive greater adoption and bring more users into the crypto ecosystem?
Catherine