Exploring how to apply GenAI to real-world financial workflows with AI Agents, RAG, and Multimodal LLMs
Welcome to the LLMs-in-Finance repository!
Here, you'll find hands-on Jupyter notebooks demonstrating how to apply the latest Generative AI tools—like OpenAI's Agents SDK, Anthropic's Claude, AutoGen, LlamaIndex, and CrewAI—to practical finance use cases.
🚀 Whether you're building a multi-agent system to analyze markets, applying RAG to earnings reports, or testing multimodal models on financial charts—this repo is for you!
If you find this work helpful, please consider giving the repository a ⭐️ to support and help others discover it!
This repo is structured into the following key areas:
- AI Agents in Finance — Use frameworks like AutoGen, LlamaIndex, and CrewAI to create collaborative agents for financial analysis.
- RAG — Learn to retrieve, parse, and analyze financial documents using Retrieval-Augmented Generation techniques.
- Multimodals — Test how well models like GPT-4o and Claude Sonnet 3.5 interpret charts in earnings reports.
- Papers in Finance — A curated summary of research papers on GenAI in financial applications.
Let’s dive in!
Below are some examples included in this repository:
- Autonomous Strategy Code Review Using AI Agents and LLM-as-a-Judge:
- Implement Financial News Bot using Evaluator-Optimizer Multi-Agent System (LLM-as-a-Judge)
-
Learn How to Configure an AutoGen Financial AI Agent! Use Case: Fetch and Analyze Data & Develop a Momentum Trading Strategy
-
How to Leverage the AutoGen Framework for Collaborative Task Management Use Case: Optimizing a Momentum Trading Strategy
- How to Get insightful information about a Stock performance in one click with code interpreter agent using LlamaIndex and Anthropic
- Looking for a simple way to integrate AI Agents into your financial analysis?
- Introspective_Agent_Worker_LLamaIndex_Financial_Tasks
- Learn how to build your own Multi-Agent Fundamental Analysis Workflow with LlamaIndex
Multi-Agent collaboration using CrewAI to:
- Provide insights of Apple' trend analysis,
- Extract sentiment analysis from news article, and
- Propose trading strategies based on these insights.
- AI-Agent For Trading Strategy Implementation: From Concept to Execution and Reflection
- How Can You Build a Financial Agentic System Using Anthropic, LlamaIndex, and APIs?
In this part, I provide several examples with various providers/libraries on how to parse complex financial reports: LlamaIndex, Anthropic, OpenAI, aisuite.
Some interesting notebooks to look at:
I've also inlcluded an example on how to evaluate your RAG system with DeepEval, GiskarAI. https://github.com/hananedupouy/LLMs-in-Finance/tree/main/RAG/evaluation
You'll find in this repo several examples showcasing the use of various multimodal LLMs such Claude sonnet 3.5, GPT-4o and recently released o1 Reasoning model. The goal is to evaluate how effectively they interpret complex charts found in financial reports.
I provide here a brief description of several papers discussing the application of GenAI in finance.
You can also find other examples related to synthetic data.
This project is intended solely for educational and research purposes.
- It is not designed for real trading or investment use.
- No warranties or guarantees are provided.
- The creator bears no responsibility for any financial losses.
By using this software, you acknowledge and agree that it is for learning purposes only.