Build Multimodal Generative AI Applications

This module explores how AI processes generate visual data by integrating images and videos with text. You will examine text-to-image/image-to-text and text-to-video/video-to-text models, image captioning, and the fusion techniques necessary for effective multimodal AI systems. Through hands-on labs, you will apply state-of-the-art models like DALL·E and Sora to generate images and videos from text prompts. Additionally, you will implement an image captioning system using Meta’s Llama 4, gaining practical experience in combining vision and language models for real-world applications.

The final module explores advanced multimodal AI applications, integrating image, text, and retrieval-based systems to build innovative solutions. You will dive into multimodal retrieval and search, multimodal Question Answering (QA), and chatbots, learning how cross-modal retrieval techniques enhance search engines and recommendation systems. Additionally, you will learn how integrating visual and textual data improves chatbot interactions. Through hands-on labs, you will build fully functional web applications with multimodal capabilities using Flask, applying state-of-the-art models and frameworks.