Skip to content

Source paper

AI-Trader: Benchmarking Autonomous Agents in Real-Time Financial Markets

Tianyu Fan, Yuhao Yang, Yangqin Jiang and 3 more

Tianyu Fan, Yuhao Yang, Yangqin Jiang, Yifei Zhang, Yuxuan Chen, Chao Huang

arXiv 2512.10971 · 20252 upvotes on Hugging Face

Research ideas from this paper (3)

Generated by AI. The paper’s authors did not write these.

AI summary

AI-Trader is a benchmarking platform designed to evaluate autonomous LLM agents in real-time, dynamic financial environments across U.S. stocks, A-shares, and cryptocurrencies. The study finds that high general intelligence does not equate to trading proficiency, as most mainstream LLMs fail to achieve positive returns or manage risk effectively without human intervention.

How it works (AI summary)

The framework utilizes a 'fully autonomous minimal information' paradigm where agents are given only basic portfolio context and a set of tools (Search, News, Math, Check Price, Trade) built on the Model Context Protocol (MCP). Agents follow a ReAct-based (Reasoning and Acting) loop to independently acquire market data, synthesize insights, and execute discrete trades without human guidance.