This chapter provides a comprehensive overview of large generative models across various data types, including vision, text, speech, image, and code. It explores the foundational architectures, key concepts, and applications that have driven advancements in Generative AI. Starting with text generative models, such as GPT and T5, the chapter delves into models that generate coherent and contextually relevant text for applications like content creation, summarization, and translation. It then moves to image and vision generative models, highlighting architectures like GANs, VAEs, and diffusion models that generate high-quality images and enable tasks such as image synthesis, style transfer, and text-to-image generation (e.g., DALL-E and Stable Diffusion). The discussion extends to speech generative models, focusing on models like WaveNet, Tacotron, and FastSpeech, which are pivotal in text-to-speech synthesis and voice cloning. The chapter also covers audio generation models, exploring how models like WaveGAN and MelGAN generate high-fidelity audio, including music and sound effects. In the realm of programming code generation, models like Codex and AlphaCode are explored for their ability to generate, complete, and translate code, thus enhancing software development workflows. Finally, the chapter examines multimodal generative models that operate across different data types, such as text-to-image or text-to-video generation, showcasing models like CLIP and Make-A-Video. This chapter offers both practitioners and research scholars a detailed understanding of the diverse landscape of generative models and their transformative applications across multiple domains.