TradingAgents is an open-source research framework in which AI agents play the staff of a trading firm and work together on stock trading decisions.
In plain words
Each agent is powered by a large language model and has one job. Analyst agents look at a company's fundamentals, market sentiment and technical signals. Bull and bear researcher agents then debate the outlook, a trader agent weighs what they found, and a risk team checks how much could be lost before a portfolio manager makes the final call. The authors' paper reports better returns than baseline models in their tests. The project says it is for research and is not trading advice.
A quick example
Imagine running it on one company's stock for a date in the past. The analyst agents each write a report, the researchers argue for and against, and the trader and risk team turn the debate into a decision you can compare with what the stock really did next.
Where you'll meet it
- Research on teams of AI agents.
- Testing AI trading ideas on past market data.
Related
Sources
- DocsTauricResearch/TradingAgents, Tauric Research (GitHub) · read 26 Oct 2026
- PaperTradingAgents: Multi-Agents LLM Financial Trading Framework, arXiv · read 26 Oct 2026
Checked by WhatIsAI against the sources above · 26 Oct 2026. Spotted an error? Report it.