LLM-X: A Scalable Negotiation-Oriented Exchange for Communication Among Personal LLM Agents

2026-05-11

AI Summary

This work addresses the lack of structured, scalable coordination mechanisms among large numbers of personal large language model (LLM) agents, as existing LLM agent communication protocols predominantly focus on API interactions. The paper proposes LLM-X, the first negotiation-oriented exchange environment, which introduces a typed message protocol to support capability negotiation and contract-net-style coordination. It further presents a system architecture comprising federated gateways, topic-based routing, and a policy enforcement engine. Empirical evaluation across 5–12 agents under diverse policies and varying short- and long-term workloads demonstrates system stability: strict policies significantly enhance robustness and fairness, albeit at the cost of increased latency and message volume, while still maintaining bounded delay drift over extended operation. This study provides the first empirical assessment of large-scale LLM multi-agent negotiation.

Abstract

We propose a personal-LLM exchange (LLM-X), a scalable negotiation-oriented environment that enables direct, structured communication across populations of personal agents (LLMs), each representing an individual user. Unlike existing tool-centric protocols that focus on agent-API interaction, LLM-X introduces a message bus and routing substrate for LLM-to-LLM coordination with guarantees around schema validity and policy enforcement. We contribute: (1) an architecture for LLM-X comprising federated gateways, topic-based routing, and policy enforcement; (2) a typed message protocol supporting capability negotiation and contract-net-style coordination; and (3) the first empirical evaluation of LLM-based multi-agent negotiation at scale. Experiments span 5, 9, and 12 agents, under distinct negotiation policies (Low, Medium, High), and across both short-run (minutes) and long-run (2h, 12h) load conditions. Results highlight clear policy-performance trade-offs: stricter policies improve robustness and fairness but increase latencies and message volume. Extended runs confirm that LLM-X remains stable under sustained load, with bounded latency drift.

Problem

• LLM agents
• scalable communication
• negotiation
• multi-agent coordination
• policy enforcement

Innovation

• LLM-X
• multi-agent negotiation
• typed message protocol
• policy enforcement
• scalable coordination