Holos: A Web-Scale LLM-Based Multi-Agent System for the Agentic Web

arXiv cs.AI / 4/6/2026

📰 NewsDeveloper Stack & InfrastructureSignals & Early TrendsModels & Research

共有:

Key Points

The paper introduces Holos, a web-scale LLM-based multi-agent system aimed at enabling a persistent “Agentic Web” where heterogeneous agents interact and co-evolve over the long term.
It argues that existing LLM multi-agent systems struggle with open-world scaling friction, coordination failures, and incentive/value dissipation, and proposes Holos to address these issues.
Holos uses a five-layer architecture centered on the Nuwa engine for efficient agent generation/hosting, a market-driven Orchestrator for more resilient coordination, and an endogenous value cycle to support incentive compatibility.
The authors position Holos as bridging micro-level agent collaboration with macro-scale emergent behavior, serving as a foundation and research testbed for future large-scale agentic ecosystems.
The system has been publicly released at https://holosai.io to provide community access and enable experimentation.

Abstract

As large language models (LLM)-driven agents transition from isolated task solvers to persistent digital entities, the emergence of the Agentic Web, an ecosystem where heterogeneous agents autonomously interact and co-evolve, marks a pivotal shift toward Artificial General Intelligence (AGI). However, LLM-based multi-agent systems (LaMAS) are hindered by open-world issues such as scaling friction, coordination breakdown, and value dissipation. To address these challenges, we introduce Holos, a web-scale LaMAS architected for long-term ecological persistence. Holos adopts a five-layer architecture, with core modules primarily featuring the Nuwa engine for high-efficiency agent generation and hosting, a market-driven Orchestrator for resilient coordination, and an endogenous value cycle to achieve incentive compatibility. By bridging the gap between micro-level collaboration and macro-scale emergence, Holos hopes to lay the foundation for the next generation of the self-organizing and continuously evolving Agentic Web. We have publicly released Holos (accessible at https://holosai.io), providing a resource for the community and a testbed for future research in large-scale agentic ecosystems.