Financial AI Agent

Description

Create you own financial AI agent using open source LLM models

How to use

1. Create a virtual environment, activate it and install all the required libraries (I used miniconda)

conda create -p venv python=3.12 # creates a virtual environment called venv in the current directory

conda activate ./venv # activate the virtual environment

pip install -r requirements.txt # install all the libraries in the requirements.txt file

2. Create an account or login to groq (not Grok) and phidata and generate API keys.

3. Save the API keys in a .env file. Make sure the .env file is in the root folder of this project

PHIDATA_API_KEY="your-phidata-api-key"
GROQ_API_KEY="your-groq-api-key"

Local Financial Agent

• You can try different models from groq website and see which suits you the best. I've used llama3-groq-8b-8192-tool-use-preview
• You can enter your prompt and accordingly change the last line of the file financial_agent.py
• Run the file financial_agent.py

python3 financial_agent.py

Financial Agent Chatbot

• Run the file playground.py. This creates a playground for your financial agent and hosts it on http://localhost:7777 by default
• Go to phidata playground section and create a endpoint
• After connecting the endpoint, you can chat with your financial agent, just like ChatGPT