AutoHedge is an open-source framework from The Swarm Corporation that assembles a team of AI agents into an autonomous hedge fund pipeline. The project has about 6,400 stars on GitHub and was updated in 2026.
Built on the Swarms multi-agent framework, AutoHedge divides trading into specialized roles. A Director agent sets the overall strategy, a Quant agent analyzes markets and signals, a Risk agent manages positions and exposure, and an Execution agent carries out the trades. The goal is to go from installation to an automated, end-to-end trading pipeline in minutes: the package installs from PyPI with a single pip command. Fully automated trading on Solana is supported today, with Coinbase support in development. Because the system trades real markets, it should be treated as experimental infrastructure for research and careful testing rather than a set-and-forget money machine. Each agent communicates through the swarm, so strategies can be inspected, adjusted, or replaced piece by piece rather than rewritten wholesale. The repository also serves as a showcase for applying the Swarms framework to a complete, real-world pipeline instead of a toy demo.
AutoHedge is free and open source under the MIT license. Trading itself involves your own capital and exchange costs, and automated trading carries real financial risk.
Quant developers and multi-agent experimenters who want a working reference architecture for agent-driven trading pipelines, and who are comfortable validating strategies before risking funds.