Text-to-image generation delves into the evolving landscape of artistic creation in the era of advanced artificial intelligence, focusing on the coexistence of human creativity and AI-generated art. We used the principles and capabilities of text-to-image generation using diffusion models, which have gained prominence in the field of artificial intelligence. Unlike traditional generative models, diffusion models employ a unique approach, starting with the destruction of training data through the addition of Gaussian noise and then learning to recover the data by reversing this process. The process aims to investigate the potential of these models to inspire new forms of artistic expression and the implications for human creativity. Therefore, we propose an optimized deep learning diffusion and Latent Consistency Model (LCM) models involving Stable Diffusion XL base 1.0 model from diffusion pipeline, LCM scheduler using Google Colab repository, torch, and datasets of text and images from COCO and Google Open Images.