Symphony: A Decentralized Multi-Agent Framework for Scalable Collective Intelligence

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Ji, Chen, Kashing, Song, Xinyuan, Zhang, Ke, Ai, Lynn, Yang, Eric, Shi, Bill
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915466523443200
author Wang, Ji
Chen, Kashing
Song, Xinyuan
Zhang, Ke
Ai, Lynn
Yang, Eric
Shi, Bill
author_facet Wang, Ji
Chen, Kashing
Song, Xinyuan
Zhang, Ke
Ai, Lynn
Yang, Eric
Shi, Bill
contents Most existing Large Language Model (LLM)-based agent frameworks rely on centralized orchestration, incurring high deployment costs, rigid communication topologies, and limited adaptability. To address these challenges, we introduce Symphony, a decentralized multi-agent system which enables lightweight LLMs on consumer-grade GPUs to coordinate. Symphony introduces three key mechanisms: (1) a decentralized ledger that records capabilities, (2) a Beacon-selection protocol for dynamic task allocation, and (3) weighted result voting based on CoTs. This design forms a privacy-saving, scalable, and fault-tolerant orchestration with low overhead. Empirically, Symphony outperforms existing baselines on reasoning benchmarks, achieving substantial accuracy gains and demonstrating robustness across models of varying capacities.
format Preprint
id arxiv_https___arxiv_org_abs_2508_20019
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Symphony: A Decentralized Multi-Agent Framework for Scalable Collective Intelligence
Wang, Ji
Chen, Kashing
Song, Xinyuan
Zhang, Ke
Ai, Lynn
Yang, Eric
Shi, Bill
Machine Learning
Artificial Intelligence
Computation and Language
Multiagent Systems
Most existing Large Language Model (LLM)-based agent frameworks rely on centralized orchestration, incurring high deployment costs, rigid communication topologies, and limited adaptability. To address these challenges, we introduce Symphony, a decentralized multi-agent system which enables lightweight LLMs on consumer-grade GPUs to coordinate. Symphony introduces three key mechanisms: (1) a decentralized ledger that records capabilities, (2) a Beacon-selection protocol for dynamic task allocation, and (3) weighted result voting based on CoTs. This design forms a privacy-saving, scalable, and fault-tolerant orchestration with low overhead. Empirically, Symphony outperforms existing baselines on reasoning benchmarks, achieving substantial accuracy gains and demonstrating robustness across models of varying capacities.
title Symphony: A Decentralized Multi-Agent Framework for Scalable Collective Intelligence
topic Machine Learning
Artificial Intelligence
Computation and Language
Multiagent Systems
url https://arxiv.org/abs/2508.20019