The Good Tech Companies - Lumoz: Decentralized Compute Infrastructure for the Era of AI, ZK, & RaaS

Episode Date: January 16, 2025

This story was originally published on HackerNoon at: https://hackernoon.com/lumoz-decentralized-compute-infrastructure-for-the-era-of-ai-zk-and-raas. This report, by Tiger Research, examines how Lumoz addresses ZKP challenges with modular infrastructure and advances AI and TEE technologies. Check more stories related to machine-learning at: https://hackernoon.com/c/machine-learning. You can also check exclusive content about #ai, #lumoz, #tee, #decentralized-computing, #zkp, #zk-implementation, #lumoz-chain, #good-company, and more. This story was written by: @lumoz. Learn more about this writer by checking @lumoz's about page, and for more stories, please visit hackernoon.com. Zero-knowledge proof (ZKP) is an emerging solution to blockchain's privacy challenges. ZKP verifies a transaction's validity without revealing details like the amount or the transactor's identity. It is a vital tool for advancing decentralized technology in line with Defensive accelerationism. Lumoz uses a modular infrastructure to gather network computing power.

This audio is presented by Hacker Noon, where anyone can learn anything about any technology. Lumos. Decentralized compute infrastructure for the era of AI, ZK, and ROS. By Lumos, formerly Opcide. This report, by Tiger Research, examines how Lumos addresses ZKP challenges with modular infrastructure and advances AI and T technologies. TLDR. ZKP promise and challenges. ZKP is a powerful blockchain technology that solves both privacy and scalability issues. However, it requires intensive computation resources, leading to high computing costs and centralization risks.

The power of decentralized modular compute layer. To solve ZKP's problems, Lumos uses a modular infrastructure to gather network computing power, accelerating ZK tech with ZK, ROS. AI and T are reshaping the future innovation. Lumos enables technology adoption beyond ZK. It provides high-performance AI computing and secures AI processes using T and DROT. These protected spaces enhance data security and result verification, driving the integration of Web3 and AI. 1. Introduction Vitalik Buterin's defensive accelerationism. D.ACC. Philosophy highlights two key principles.

It accelerates technological progress and resists centralized power. It advocates for decentralized development. Blockchain technology aligns with this philosophy. It decentralizes power and prevents centralization. However, blockchain technology has trade-offs. Decentralized systems operates lower as the network grows. The consensus process limits performance. Data storage increases exponentially and restricts scalability. On-chain transparency ensures accountability but reduces privacy. Zero-knowledge proof, ZKP, is an emerging solution to blockchain's privacy challenges. This technology lets users

prove something is true without sharing the actual information. Like proving you're old enough to vote without revealing your birth date it's a way to verify facts while keeping sensitive details private when used in blockchain zero knowledge proofs allow the network to verify that transactions are valid without exposing private details like the amount transferred or who sent it this lets users keep their information private while still ensuring the system remains trustworthy in In blockchain, ZKP technology solves two key challenges. The first is privacy. ZKP verifies a transaction's validity without revealing details like the amount or the transactor's identity. The second is scalability.

The blockchain can process computations externally and use ZKP to validate the results. This reduces the network's computational load and improves efficiency zkp works well in layer 2 scaling solutions layer 2 processes transactions off chain and submits the results to the ethereum mainnet with proof of correct execution this improves processing speed and scalability while maintaining system integrity zkp addresses both privacy and scalability issues. It is avatar tool for advancing decentralized technology in line with defensive accelerationism. 2. Challenges in ZK implementation. High costs and centralization risks,

generating proofs with ZKP technology is computationally intensive. It requires complex cryptographic operations and computing power similar to Bitcoin Mininger training large AI models. These high-performance computing requirements are especially evident in Layer 2 scaling solutions like ZK rollups. For instance, generating a ZKP for a batch of 250 ERC-20 transactions on Polygon ZKE VM takes about 2 minutes using a server with AMD 224 vCPUs and 896GB RAM. The operational cost is approximately $0.32 for 2 minutes or $6,909 monthly. In contrast, a typical web server requires only 4 vCPUs and 8GB RAM. These high hardware requirements create two major limitations for

ZKP technology. First, only a few high-powered machines can create these proofs, which goes against the goal of spreading out control among many users. This reliance makes it hard to achieve true decentralization. Second, computing resources restrict the size of proof batches that can be processed. This reduces the efficiency and scalability of ZKP technology. To overcome these limitations, ZKP infrastructure must become more efficient and accessible. Reducing technical complexity and lowering barriers to entry are key to scaling the ZKP ecosystem. Lumos uses its modular computing infrastructure and decentralized prover network to address these issues. It aims to help solve the blockchain trilemma of decentralization, scalability, and security. 3. LUMOZ

Illuminating the path to ZK, ACC, Lumos is a modular computing layer for ZK, AI, and ROS. It offers robust computing power and verification services for ZK, AI, and ROS. It offers robust computing power and verification services for ZK and AI applications across chains with diverse architectures. Network participants share computing power to reduce centralized dependencies and achieve true decentralization. Also, Lumos' ZK-ROS platform lowers barriers for ZK rollups and accelerates ZK technology through ZK, ACC. Lumos's technology is recognized in the market. More than 20 Layer 2 chains, including Carve, Merlin Chain, ZK Fair, Ultiverse, Matter1X, and Uxlink, use the Lumos infrastructure. In early 2024, Lumos raised $14 million in funding. Leading investors like OKX Ventures,

Hashkey Capital, KuCoin Ventures, and Polygon supported the funding. 3. 1. LUMOZ Chain, ZK Modular Computing Layer Lumos Chain will launch its mainnet in Q1 2025. It aims to create a centralized physical infrastructure network, DEPIN, for high-performance computational tasks, including ZKP technology. Lumos offers a 1. modular architecture, 2. a hybrid consensus model, and 3. optimized performance for intensive computation. This should make zero-knowledge proofs more affordable and accessible while helping blockchain networks handle more users without relying on central control. 3. 1. 1. Modular architecture Lumos is built with a modular design that lets each project customize its setup. Think of it like building with blocks, projects can pick and arrange the pieces they need. Projects can configure their infrastructure based on specific

needs. Each module operates independently to ensure efficient system configuration. This approach reduces integration complexity and improves overall system efficiency. Lumos's modular structure offers limitless applications. For example, it can function as a prover layer to generate proofs for key ZK rollups like Polygonsky VM, ZKSYNC, and Scroll. It can also serve as a compute layer to handle AI model training and inference. This design ensures flexibility and supports the integration of new technologies and services. 3. 1. 2. Hybrid Consensus Mechanism Lumos uses a hybrid structure to manage the generation and verification of ZKPs. At its core are two components,

ZKProver, which generates ZKPs, and ZKVerifier, which verifies them. ZKProver uses a proof-of-work POW consensus structure. ZKP generation requires high-performance computation. Any individual or organization with GPU, CPU resources can join. They use their computing power to generate ZKPs and perform other tasks. ZK Verifier operates as a verification network based on proof-of-stake, POS. It consists of 100 K ZK Verifier nodes issued through the sale of node licenses. License holders can either run these nodes as dedicated clients or delegate them to others. The ZK verifier ensures the network's trustworthiness by validating the proofs generated by ZK prover. This hybrid structure uses a consensus mechanism suited

to each component's role. POW handles high-performance computations for ZKP generation. POS manages verification tasks. This approach reduces reliance on centralized, high-performance equipment. It ensures efficient verification through a distributed network of provers. As a result, proof generation and verification, the core elements of ZKP, are handled efficiently in a decentralized manner. 3. 1. 3. Optimized performance LUMOS consistently enhances performance through ongoing research and diverse technological advancements. Notably, it has optimized the processes of proof generation and verification. LUMOS enhances ZKP generation and verification with an aggregator service.

This service processes multiple proofs in parallel and combines them into a single proof. Traditional methods handle each proof independently, creating inefficiencies from redundant calculations. Lumos removes this bottleneck and significantly increases processing speed. Lumos also optimizes the verification process. Previously, proving required two separate steps. Provers submitted a proof hash with their identity and work information in the first step and their ZKP in the second step. Lumos simplifies this by combining identity, work information, and ZKP into a single contract. This reduces on-chain transactions by half, cutting gas costs by 50-60% and shortening verification time. Lumos further optimized the proof generation process.

First, it breaks large tasks into smaller pieces that can be worked on separately. Then, when parts of different calculations are the same, Lumos saves and reuses the cached results instead of calculating them again. Additionally, when computers in the network aren't busy, they prepare calculations that might be needed soon, rather than waiting for new tasks to arrive. These changes increased proof generation speed by 25%. They also improved cost efficiency and processing performance significantly. 3. 2. ZK-ROS platform Lumos offers a ROS service that helps developers build ZK rollup-based Layer 2 chains without needing complex infrastructure. This service

simplifies the process, much like AWS simplifies cloud computing. ROS provides the necessary infrastructure to build and operate chains with rollup technology. Its intuitive interface simplifies development and operations. This improves user convenience and efficiency. Lumos's core strength comes from its proprietary network. This network includes a decentralized computing layer, ZK prover, and ZK verifier. It is specially designed to handle the complex calculations needed for ZK rollups, a way to process many blockchain transactions quickly. This capability distinguishes Lumos from other raw services without dedicated computing networks.

The decentralized computing layer serves as a unique moat for Lumos. ITRI enforces its competitive advantage in the roll-up ecosystem. Lumos offers an environment that connects and extends various chains in ZK roll-up. Developers can build custom chains without dealing with complicated technical setups, making it much simpler to design and run their systems. Lumos Rollup Launchbase supports major chains like Bitcoin, Ethereum, BNB Chain, Polygon, and Solana. It also supports ZKE VM types such as Polygon ZKE VM, ZKSYNC, and Scroll. This wide support makes building ZK rollup solutions easier for developers.

Lumos also enables native cross-rollup communication, NCRC, for efficient crypto transfers and interactions across rollups. This feature uses the Rollup System Contract, RSC, deployed on major chains. The RSC keeps track of information and updates between different rollups, allowing for quick and secure transfers and interactions while keeping all assets safe. By the first half of 2025, Lumos plans to support major L1 chain. It will also expand to support EVM-based ZK rollups, including Solana Virtual Machine, SVM, and Ton Virtual Machine, TVM. This expansion meets the rising demand for ZK

technology and strengthens Lumos's position as a key provider of ZK infrastructure. 4. Lumos boosts future vision. Lumos's network of distributed computers not only handles complex calculations efficiently but also can grow to meet increased demand while staying reliable. While it's currently focused on zero knowledge proofs, this powerful infrastructure could help develop other advanced technologies in the future. 4.1. ETH 3.0. Make Ethereum Great Again The vision of Ethereum 3.0 is to enhance the network's scalability, security, and efficiency through snarkification and beamchain. Snarkification speeds up transaction

processing and improves data security by introducing ZKP technology into Ethereum's consensus and execution layers. To achieve this, powerful computing resources are required. Lumos provides this through its modular computing layer. It uses a hybrid consensus mechanism combining POW and POS. POW delivers the computing power for ZKP generation, while POS handles verification tasks. This setup enables high-performance computation in a decentralized manner. This infrastructure helps Lumos optimize validators and speed up block generation, key goals of Beamchain. The network can improve processing efficiency while maintaining decentralization.

By supporting these improvements, Lumos plays an important role in helping Ethereum grow and advance. 4.2. Build Computing Infrastructure For AIERA Lumos will serve as a key infrastructure for advancing AI technology. Right now, most AI systems are controlled by large technology companies. Lumos's distributed network of computers aims to change this by making AI more accessible to everyone. This could help create an AI ecosystem that gives users more control. Makes AI technology easier and cheaper to access. Operates more fairly and openly. This approach could lead to a new kind of AI system that isn't controlled by just a few companies. Lumos uses two technologies to make AI systems more secure and trustworthy.

Decentralized Root of Trust, DROD, using Distributed Computing and Trusted Execution Environment, T. AI agents act autonomously, but most currently operate in untrusted environments, limiting their security and transparency. In these environments, protecting sensitive information and verifying results is difficult, reducing decision reliability. DRAWT system keeps important security keys protected in hardware and uses a verification process called remote attestation to ensure everything works as intended. This boosts data security and execution integrity while improving result

verifiability. With these security measures in place, AI agents can work more safely and reliably. This brings together the best of Web3 and AI technologies, helping create a next-generation AI system that is both secure and transparent. 5. Closing Thoughts Lumos has already shown significant progress, with its test network drawing over 3 million users and supporting more than 50 major ecosystem projects. It serves over 20 L2 chains and continues expanding its ecosystem through partnerships with UXLINK, CARV, MerlinChain, Matter1X, Ultiverse, ZKFair, and others. Lumos goes beyond its current successes to tackle a fundamental challenge in Web3. By using zero-knowledge-proof technology and distributed computing,

it makes blockchain networks both faster and more transparent. The platform's work with artificial intelligence opens up new growth opportunities. This approach stays true to Web3's core principle of distributed control while showing how combining ZK technology with AI can help the entire ecosystem advance. As new technologies like Ethereum's Beam Chain require more computing power for zero-knowledge proofs, the demand for processing power is expected to grow dramatically. This increase might be similar to how Bitcoin and Ethereum mining needed more and more powerful computers in their early days. By providing the essential infrastructure to handle these growing computing needs, Lumos is well positioned to become a crucial part of both blockchain and

AI development. Lumos also supports various technical environments, including Trusted Execution Environments and Fully Homomorph, FHEs. These tools help build a stronger foundation for the future of blockchain and AI development. Given how Lumos combines these capabilities with its other features, its growth and impact on the industry will be important to watch. Disclaimer, this report was partially funded by Lumos. It was independently produced by our researchers using credible sources. The findings, recommendations, and opinions are based on information available at publication time and may change without notice. We disclaim liability for any losses from using this report or its contents and do not warrant its accuracy or completeness. The information may differ from

others' views. This report is for informational purposes only and is not legal, business, investment, or tax advice. Referencesto securities or digital assets are for illustration only, not investment advice or offers. This material is not intended for investors. Terms of usage Tiger Research allows the fair use of its reports. Fair use is a principle that broadly permits the use of specific content for public interest purposes, as long as it doesn't harm the commercial value of the material. If the use aligns with the purpose of fair use, the reports can be utilized without prior permission. However, when citing Tiger Research's reports, it is mandatory to 1. Clearly state

Tiger Research as the source, 2. Include the Tiger Research logo, black, white. If the material is to be restructured and published, separate negotiations are required. Unauthorized use of the reports may result in legal action. Thank you for listening to this HackerNoon story, read by Artificial Intelligence. Visit HackerNoon.com to read, write, learn and publish.