README

FinRobot: An Open-Source AI Agent Platform for Financial Applications using Large Language Models





!License
!
!
!
!






FinRobot is an open-source agentic AI framework for financial decision intelligence. It combines foundation models with financial tools, quantitative models, deterministic computation, and multi-agent workflows to build decision-grade financial applications.

Its core principle is simple: models reason, software computes, agents orchestrate, and systems verify. Models can be replaced; the surrounding financial infrastructure — tools, workflows, validation, provenance, and evaluation — is what makes the system reliable.

Unlike the single-model paradigm of FinGPT, FinRobot supports end-to-end agentic workflows for equity research, quantitative trading, risk analysis, investment banking, private equity, wealth management, and other high-stakes financial decisions.

Whitepaper on arXiv · Official Academic Page

!Visitors


---

 Where to start

FinRobot is three projects sharing one financial domain layer. They are not three versions of the same thing competing for your attention — they serve different purposes.

| Directory | Version | What it is | Use it for |
|:---|:---|:---|:---|
| finrobot_desktop/ | V2 | Production — the real agent system. Native desktop research workstation on PydanticAI + FastAPI + React/Tauri | Actual research work. This is the one to use if you want output you can act on |
| finrobot_equity/ | V1 | Web app — a self-hosted report generator with a FastAPI interface | Standing up a browser-based service that turns a ticker into a shareable HTML/PDF report |
| finrobot_autogen/ | V0 | Educational — the original AutoGen library, the codebase behind the whitepaper | Learning how financial agents are wired together, teaching, and reproducing the paper. Not intended for production use |

pip install finrobot installs the V0 package. Its source moved into finrobot_autogen/, but the import name did not change — every existing from finrobot... import ... keeps working.

---

 🧬 Architecture Evolution

FinRobot evolves alongside the rapid development of AI-agent frameworks. Rather than being tied to a single agent stack, each generation explores how emerging agent architectures can improve financial analysis, research, and decision-making.

| Version | Agent Framework | Project | Maturity | Focus |
|---|---|---|---|---|
| V0 | AutoGen | finrobot_autogen/ | Educational / reference | The original FinRobot multi-agent architecture for financial applications |
| V1 | OpenAI Agents SDK | finrobot_equity/ | Self-hosted web app | Equity-research agents, financial analysis, valuation, and automated report generation |
| V2 | PydanticAI | finrobot_desktop/ | Production — also hosted at finrobot.ai/v2 | Structured, type-safe agent workflows for professional equity research |
| V3 | DeepSeek-Harness | FinRobot V3 | In development | More autonomy inside the same verification constraints |

All three are open source. V2 is