Interacting with Multimodal models in GitHub Models and Microsoft Copilot

Innovate with Azure OpenAI's GPT-4o multimodal model in this hands-on experience in GitHub Models and Microsoft Copilot. Learn the core concepts and best practices to effectively generate with text, sound, and images using GPT-4o-mini and Copilot.

This workshop is for the Global AI Bootcamp 2025

Description

This workshop is designed to give you a hands-on introduction to the core concepts and best practices for interacting with OpenAI models in GitHub Models portal. We will be running the workshop using GitHub Models and Microsoft Copilot.

Duration

45 - 60 minutes

Slide Deck

Download the slidedeck here.

Learning Outcomes

  • Understand how Large Language Models work, including what tokens are​.
  • Explore Prompt Engineering techniques and best practices​
  • Understand how models apply existing knowledge​

Technology Used

  • GitHub Models
  • Microsoft Copilot

Learning Objectives

By the end of this workshop, you should be able to:

  • Describe GitHub Models capabilities and implementation in the fields of text generation, image generation and multimodal conversation.
  • Understand prompt engineering concepts and best practices.
  • Leverage generative AI capabilities and apply prompt engineering techniques to your own solutions.

Lab Outline

The lab is organized into 3 sections, taking you through generating text content, image assets, and code snippets through a multimodal conversational interface on GitHub models. The goal of the lab is to leverage generative AI capabilities over a wide range of scenarios.

  1. Part 1 - Text Generation Generate text content and descriptions with GPT4o
  2. Part 2 - Image Generation Generate image assets with DALLE-3
  3. Part 3 - Multimodality Leverage multimodal capabilities of GPT-4o to interact with images and text.

Move to Setup to set up your Workshop environment and get started.