TradingAgents: Multi-Agents LLM Financial Trading Framework
Summary
This paper introduces TradingAgents, a multi-agent LLM framework that simulates real-world trading firms to improve stock trading performance. It utilizes specialized agents for analysis and risk management, demonstrating superior results in cumulative returns and Sharpe ratio compared to baselines.
View Cached Full Text
Cached at: 05/08/26, 08:32 AM
Paper page - TradingAgents: Multi-Agents LLM Financial Trading Framework
Source: https://huggingface.co/papers/2412.20138 Published on Dec 28, 2024
Abstract
A multi-agent framework using large language models for stock trading simulates real-world trading firms, improving performance metrics like cumulative returns and Sharpe ratio.
Significant progress has been made in automated problem-solving using societies of agents powered bylarge language models(LLMs). In finance, efforts have largely focused on single-agent systems handling specific tasks or multi-agent frameworks independently gathering data. However, the multi-agent systems’ potential to replicate real-world trading firms’ collaborative dynamics remains underexplored. TradingAgents proposes a novel stock trading framework inspired by trading firms, featuringLLM-powered agents in specialized roles such asfundamental analysts,sentiment analysts, technical analysts, andtraderswith varied risk profiles. The framework includes Bull and Bear researcher agents assessing market conditions, arisk managementteam monitoring exposure, andtraderssynthesizing insights from debates and historical data to make informed decisions. By simulating a dynamic, collaborative trading environment, this framework aims to improve trading performance. Detailed architecture and extensive experiments reveal its superiority over baseline models, with notable improvements in cumulative returns,Sharpe ratio, andmaximum drawdown, highlighting the potential of multi-agentLLMframeworks in financial trading. TradingAgents is available at https://github.com/TauricResearch/TradingAgents.
View arXiv pageView PDFGitHub71.3kautoAdd to collection
Get this paper in your agent:
hf papers read 2412\.20138
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2412.20138 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2412.20138 in a dataset README.md to link it from this page.
Spaces citing this paper4
Collections including this paper37
Similar Articles
TauricResearch/TradingAgents
TradingAgents is an open-source multi-agent LLM framework for financial trading, with support for various LLM providers and recent updates including new models and features.
Agentic Trading: When LLM Agents Meet Financial Markets
This paper presents a systematic survey and evidence map of 77 studies on LLM-based trading agents, finding that architectural experimentation is expanding rapidly but evaluation protocols, execution semantics, and reproducibility remain critical bottlenecks.
QuantAgent: Price-Driven Multi-Agent LLMs for High-Frequency Trading
QuantAgent is a multi-agent LLM framework designed specifically for high-frequency trading, using four specialized agents (Indicator, Pattern, Trend, Risk) to make rapid, risk-aware decisions based on short-horizon signals. In zero-shot evaluations across ten financial instruments including Bitcoin and Nasdaq futures, it outperforms existing neural and rule-based baselines in predictive accuracy and cumulative return.
@DanKornas: A single LLM can blur market data, debate, and risk into one opaque answer. This repo splits the work across specialize…
TradingAgents is an open-source multi-agent LLM framework for financial market analysis, routing market data through specialized agents for research.
Beyond Agent Architecture: Execution Assumptions and Reproducibility in LLM-Based Trading Systems
This paper reviews and audits execution realism in LLM-based trading research, proposing clearer reporting standards for reproducibility and evaluation comparability.