ai4finance-foundation/finrobot

★ 6,949Jupyter NotebookAudience · dataComplexity · 3/5Setup · moderate

Mindmap

mindmap
  root((FinRobot))
    What It Does
      Equity research reports
      Market forecasting
      Trading strategy
    AI Pipeline
      Multi-agent system
      Swappable LLM backends
    Data Sources
      Financial APIs
      Income statements
    Output Formats
      HTML reports
      PDF documents
    Setup
      API keys needed
      Web interface

mindmap root((FinRobot)) What It Does Equity research reports Market forecasting Trading strategy AI Pipeline Multi-agent system Swappable LLM backends Data Sources Financial APIs Income statements Output Formats HTML reports PDF documents Setup API keys needed Web interface

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Generate a multi-page equity research report with valuation and risk sections for any company ticker in minutes.

USE CASE 2

Swap in different language models to compare analysis quality across AI providers for financial research.

USE CASE 3

Run the web interface to let non-technical team members trigger financial reports without code.

USE CASE 4

Use the document analysis tools to quickly extract key figures from earnings releases or SEC filings.

Tech stack

PythonJupyter NotebookOpenAI

Getting it running

Difficulty · moderate Time to first run · 1h+

Requires API keys for both a financial data provider and an AI model provider (OpenAI by default), configured in a local config file.

Released as open source by the AI4Finance Foundation, the specific license terms are not described in the explanation.

In plain English

FinRobot is an open-source platform that uses AI language models to automate financial analysis tasks. Its primary output is professional equity research reports: given a company ticker symbol, it fetches financial statements, runs projections, and produces a multi-page HTML or PDF document with charts covering valuation, peer comparison, cash flow, and investment risk. The platform is aimed at investors, analysts, and researchers who want that kind of structured analysis without assembling it manually. The system works through a pipeline. It first pulls income statements, balance sheets, and cash flow data from a financial data API. It then runs those numbers through several specialized AI agents, each focused on a different aspect of the company (investment thesis, risk assessment, valuation, and so on). Finally it combines the agents' outputs into a formatted report. Example reports for companies like NVIDIA and Microsoft are linked in the README. Beyond equity reports, the platform includes tools for market forecasting, document analysis, and trading strategy development. It is built to support multiple AI model backends, so researchers can swap in different language models depending on the task. The project also ships a web interface that can be started with a single script, as well as command-line tools for running analysis steps individually. Setting up the project requires API keys for a financial data provider and an AI model provider (OpenAI by default), both of which are configured in a local config file. The codebase is primarily Python, with many Jupyter notebooks that walk through specific use cases and demonstrations. FinRobot is released under an open-source license by the AI4Finance Foundation, the same group behind the related FinGPT project. A hosted version called FinRobot Pro is also available at a separate website for users who prefer not to run it locally.

Copy-paste prompts

Prompt 1

I want to generate an equity research report for NVIDIA using FinRobot. Walk me through setting up the config file with my OpenAI and financial data API keys and running my first report.

Prompt 2

How do I swap the default OpenAI backend in FinRobot for a different language model provider? Show me which config values to change.

Prompt 3

I want to add a new specialized agent to FinRobot focused on ESG risk assessment. Where in the pipeline code would I add it?

Prompt 4

Show me how to run FinRobot's market forecasting tools for a specific sector and export the results to a CSV.

Prompt 5

Help me set up FinRobot's web interface so my team can generate financial reports without touching the command line.

Open on GitHub → Explain another repo

← ai4finance-foundation on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.