Simple Azure AI Streamlit Chat Template

This project provides a simple Streamlit chat app to get you started with the Azure AI Foundry and Azure AI Inference client library for Python.

📝 Overview

Features

• Supports both text and image inputs
• Seamlessly switch between preconfigured AI models via a dropdown menu
• Select from predefined system prompt templates or enter custom instructions
• Non-persistent chat history managed with Streamlit session state, including options to view or clear the session
• Easily clear the chat history at any time
• Print messages and session state to console for troubleshooting

Azure AI Inference client library

This project is using the Azure AI Inference client library for Python (azure-ai-inference). If preferred, you can modify the code to use the OpenAI Python client library (openai) instead.

For a list of supported models, services, and known issues, refer to the Azure AI Inference client library for Python documentation.

🚀 Quick Start

Use GitHub Codespaces, Dev Containers in VS Code, or setup manually:

1. Clone the repository:

git clone https://github.com/gerbermarco/simple-azure-ai-streamlit-chat.git
cd simple-azure-ai-streamlit-chat

2. Copy and modify the .env file:

cp .env.example .env
# Edit .env and update the values with your details

Note: Never commit your .env file to version control, as it contains sensitive information.

3. Install dependencies:

pip install -r requirements.txt

4. Run the Streamlit app:

streamlit run app.py