Repolode
Submit
virattt

virattt/ai-hedge-fund

An educational proof of concept that uses multiple LLM-powered investor agents to explore AI-driven trading decisions; it does not execute trades.

View on GitHub
Stars
64,532
Forks
11,343
Open issues
187
Last commit
4 days ago

Repolode describes open-source projects and does not give investment advice. Trading and investing can lose money: assess the risk yourself and follow the laws where you live before using this project.

What is it?

AI Hedge Fund is an experimental project that explores whether AI agents can collaborate to make trading decisions. It is built for educational and research purposes, and the system does not place any real trades or provide investment advice.

The project ships as a Python package, aihf, that launches an interactive terminal app. Users can run paper trading funds that record each session in a hash-chained local ledger, or backtest a fund over historical data and compare its equity curve against a benchmark. API keys are requested on first use and stored in a local .env file.

Market data comes from the Financial Datasets API, either with a key or via x402 micropayments in USDC on Base. Investor agents can be backed by Anthropic, OpenAI, DeepSeek, Google, xAI, Kimi, or TypeSafe models.

Key features

  • 01Interactive terminal app launched with the aihf command
  • 02Paper trading funds with a per-session, hash-chained local ledger
  • 03Backtesting with equity curves compared against a benchmark
  • 04Support for multiple LLM providers for the investor agents
  • 05Market data via Financial Datasets API, with a key or x402 payments

Use cases

  • Studying how multiple LLM agents with different investing styles reason about stocks
  • Backtesting a multi-agent strategy against historical market data
  • Experimenting with paper trading workflows without risking real money

Quick start

  1. 1

    Install the app with pipx

    pipx install aihf
  2. 2

    Launch the interactive app from anywhere; it will ask for API keys on first use

    aihf

Things to know

  • Educational proof of concept only; it does not execute trades and is not meant for real investing.
  • Requires a Financial Datasets API key (or x402 USDC payments) plus a model API key from a supported provider.
  • Backtesting and paper trading use last 365 days of data for x402 non-professional use.

In collections

Open-source AI agent frameworks & apps

Topics

#ai#llm#trading#backtesting#agents#python#finance#paper-trading

FAQ

What is ai-hedge-fund?
An educational proof of concept that uses multiple LLM-powered investor agents to explore AI-driven trading decisions; it does not execute trades. AI Hedge Fund is an experimental project that explores whether AI agents can collaborate to make trading decisions. It is built for educational and research purposes, and the system does not place any real trades or provide investment advice.
Is ai-hedge-fund free to use? What license does it have?
ai-hedge-fund is open source under the MIT license. Check the license terms before using it in commercial products.
Is ai-hedge-fund still maintained?
Actively maintained. The last commit to ai-hedge-fund was on Nov 3, 2026.
What can I use ai-hedge-fund for?
Studying how multiple LLM agents with different investing styles reason about stocks Backtesting a multi-agent strategy against historical market data Experimenting with paper trading workflows without risking real money
How do I install ai-hedge-fund?
Install the app with pipx: pipx install aihf Launch the interactive app from anywhere; it will ask for API keys on first use: aihf
Where can I get ai-hedge-fund?
The source code and installation instructions are on GitHub: https://github.com/virattt/ai-hedge-fund

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects