Skip to content
Sign in
  • Browse all toolsFilter by category and pricing
  • Browse categoriesWriting, code, image, audio and more
  • Compare toolsPut up to three side by side
  • Read the blogGuides and comparisons

FinRobot

Open-source financial AI agents for equity research, code-computed valuations, and traceable investment reports.

FinRobot is a financial AI agent framework that combines language models, deterministic calculations, and multi-agent workflows for equity research and investment analysis. FinRobot serves financial analysts, developers, and educators through desktop, browser, and command-line workflows, with source code available under the Apache-2.0 license.

Updated 9 Nov 20266 sections

FinRobot website screenshot

What is FinRobot?

FinRobot is an open-source agentic AI framework for financial decision intelligence, published in the AI4Finance-Foundation repository. FinRobot combines foundation models with financial tools, quantitative models, deterministic computation, and multi-agent workflows. The project supports financial applications spanning equity research, quantitative trading, risk analysis, investment banking, private equity, and wealth management.

FinRobot separates the responsibilities of models, software, and agents. Language models reason and explain, software performs financial calculations, and agents coordinate workflows. The repository contains a production desktop research system, a self-hosted equity research web application, and an educational AutoGen library, each serving a different purpose rather than acting as interchangeable installations.

What FinRobot can do

FinRobot combines research orchestration with code-based financial analysis. The production system takes market data and company filings through analysis, modeling, synthesis, reporting, and investment debate. Financial numbers come from pure-Python operators rather than language-model generation, while language models interpret the results and write accompanying explanations.

  • FinRobot calculates DCF, DDM, LBO, WACC, comparable-company analysis, and Monte Carlo outputs.
  • FinRobot runs company research, earnings analysis, and investment committee memo pipelines.
  • FinRobot coordinates specialized data, analysis, modeling, synthesis, and reporting agents.
  • FinRobot evaluates investment arguments through bull, bear, and judge agents.
  • FinRobot tracks provenance for outputs and includes audit operators.
  • FinRobot generates shareable HTML and PDF reports through the equity research web app.

How FinRobot works

FinRobot offers the production research engine through a native desktop application, a local web interface, or a CLI. The desktop application uses PydanticAI, FastAPI, and React/Tauri. The browser workflow runs a FastAPI backend alongside a Vite frontend and supports Intel Macs, Linux, and Windows without requiring the desktop installer or Rust toolchain.

FinRobot routes research requests through a lead agent and specialized pipeline agents before producing a traceable research output. Data infrastructure includes FMP, Finnhub, yfinance, SEC EDGAR, Adanos, NewsAggregator, and FX, with provider failover. CLI examples cover company research, DCF valuations, comparable-company analysis, investment committee memos, and questions about company-specific financial exposure.

Who FinRobot is for

FinRobot serves analysts performing equity research and financial decision workflows, developers building financial applications, and educators studying agent architectures. The repository recommends the production system for actual research work. The original AutoGen library instead supports learning, teaching, and reproducing the whitepaper, with tutorials covering annual reports, forecasting, strategy development, and chart analysis.

  • FinRobot supports analysts researching companies and preparing investment committee reports.
  • FinRobot supports teams deploying browser-based equity research report services.
  • FinRobot provides financial agent tutorials for developers and educators.

FinRobot pricing and source access

FinRobot is an open-source project distributed under the Apache-2.0 license. The repository provides source code for the production desktop system, the equity research web app, and the educational AutoGen framework. Access options include a downloadable desktop release, local installation instructions, and a hosted production system referenced at finrobot.ai/v2.

FinRobot does not list subscription prices or hosted service plan details in the supplied repository text. The equity research web app requires configuration of FMP and OpenAI API keys, with an optional Adanos key. Installing the finrobot package through pip installs the educational AutoGen library rather than the production desktop research system.

FinRobot strengths and limitations

FinRobot provides a defined separation between financial computation and language-model narration, supported by provenance tracking and audit operators. Multiple interfaces allow the production engine to support desktop, browser, and command-line workflows. Deployment constraints remain important: the native release targets Apple Silicon, browser mode exposes an unauthenticated local API, and the educational package is not intended for production.

  • FinRobot has no Intel Mac desktop build in the documented release.
  • FinRobot is not yet Apple-notarized and may require a Terminal command to launch.
  • FinRobot requires dependency installation for the local web workflow.
  • FinRobot requires external API configuration for the equity research web app.

FinRobot FAQ

What is FinRobot?

FinRobot is an open-source agentic AI framework for financial decision intelligence. FinRobot combines language models, financial tools, deterministic calculations, and multi-agent workflows for applications including equity research, valuation, risk analysis, investment banking, and wealth management.

Is FinRobot free and open source?

FinRobot is available as open-source code under the Apache-2.0 license. The repository includes desktop, self-hosted web, and educational projects, but does not state hosted service prices or subscription plans. The equity research web app requires external API keys.

Does FinRobot calculate financial numbers with a language model?

FinRobot calculates financial numbers through deterministic pure-Python operators rather than language-model generation. Language models handle reasoning, synthesis, explanation, and report writing, while code computes valuation outputs such as DCF, DDM, LBO, WACC, comparable-company analysis, and Monte Carlo simulation.

Can FinRobot run on Windows or Linux?

FinRobot supports Windows and Linux through the production engine's local web interface. The same browser workflow also works on Intel Macs without a Rust toolchain. The documented native desktop download supports macOS Apple Silicon, with no Intel Mac build available.

Does pip install finrobot install the production application?

FinRobot installs the educational AutoGen library when users run pip install finrobot. The repository identifies that library as a reference for learning, teaching, and reproducing the whitepaper, not production research. The production system is provided separately in finrobot_desktop.

Which financial data sources does FinRobot support?

FinRobot lists FMP, Finnhub, yfinance, SEC EDGAR, Adanos, NewsAggregator, and FX in the production system's data infrastructure, with provider failover. The separate equity research web app requires FMP and OpenAI keys and optionally accepts an Adanos key.

From the community

No reviews yet

Be the first to review this tool. What you write helps the next person choose.

More tools like FinRobot

StockInsights

www.stockinsights.ai

AI equity research for searching company filings, summarising earnings calls and monitoring disclosures.

Freemium

AI Berkshire

www.github.com

Open-source value investing research skills for Claude Code and Codex.

Free

TradingAgents

www.github.com

Open-source financial trading research framework with LLM analysts, researcher debates, and portfolio risk assessment.

Free