PolyLink: A Blockchain Based Decentralized Edge AI Platform for LLM Inference

2025-10-01

AI Summary

To address trust deficits and high operational costs arising from the centralized architecture of current LLM services, this paper proposes the first blockchain-based decentralized edge AI platform enabling distributed inference and collaborative development of large language models across heterogeneous edge devices. Our method introduces three core innovations: (1) the TIQE protocol, ensuring end-to-end inference integrity; (2) a lightweight cross-encoder combined with an LLM-as-a-Judge mechanism for efficient, high-accuracy result verification; and (3) a token-incentivized economic model integrating dynamic pricing to enhance security and participant engagement. Experimental evaluation demonstrates that the system achieves bounded inference latency, robustness against model degradation attacks, and resilience to verifier tampering—across geographically distributed edge infrastructure. The platform is fully open-sourced and has undergone end-to-end validation.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Security and Privacy: Blockchains and distributed ledgers

Abstract

The rapid advancement of large language models (LLMs) in recent years has revolutionized the AI landscape. However, the deployment model and usage of LLM services remain highly centralized, creating significant trust issues and costs for end users and developers. To address these issues, we propose PolyLink, a blockchain-based decentralized AI platform that decentralizes LLM development and inference. Specifically, PolyLink introduces a decentralized crowdsourcing architecture that supports single-device and cross-device model deployment and inference across heterogeneous devices at the edge. Moreover, to ensure the inference integrity, we design the TIQE protocol, which combines a lightweight cross-encoder model and an LLM-as-a-Judge for a high-accuracy inference evaluation. Lastly, we integrate a comprehensive token-based incentive model with dynamic pricing and reward mechanisms for all participants. We have deployed PolyLink and conducted an extensive real-world evaluation through geo-distributed deployment across heterogeneous devices. Results indicate that the inference and verification latency is practical. Our security analysis demonstrates that the system is resistant to model degradation attacks and validator corruptions. PolyLink is now available at https://github.com/IMCL-PolyLink/PolyLink.

Problem

• Decentralizing LLM development and inference services
• Ensuring integrity of AI inference through verification protocols
• Creating incentive mechanisms for distributed edge computing participants

Innovation

• Blockchain-based decentralized edge AI platform
• TIQE protocol ensures inference integrity verification
• Token incentive model with dynamic pricing mechanisms