Skip to main content
Back to Research
Multi-Agent Systems

TradingAgents: Multi-Agents LLM Financial Trading Framework

Xiao, Y. et al. · 2024

What this means for traders

Having AI agents debate and challenge each other reduces errors and improves risk-adjusted returns compared to using a single AI model.

Abstract

Explores the debate mechanism between opposing agents. The study finds that agent debate reduces hallucinations and improves risk-adjusted returns (Sortino/Sharpe ratios) compared to single-agent models.

Multi-AgentDebateRisk ManagementTrading Framework
Read the full paper