Text-to-image generation has become an important research area in artificial intelligence and deep learning.Recent advancements in diffusion models have significantly improved the ability of machines to generate highquality images from textual descriptions. This paper presents a text to image generation system implementedusing diffusion models and CLIP text encoders. The system takes a natural language prompt as input andgradually transforms random noise into a meaningful image through iterative denoising steps. Theimplementation is performed using Python and PyTorch within Visual Studio Code and Jupyter Notebookenvironments. Experimental outputs demonstrate that diffusion models are capable of generating visuallycoherent images that align with textual prompts. The study highlights the effectiveness of diffusion basedgenerative models for creative applications such as digital art generation, automated media production, anddesign assistance.