Consistency Models (CMs) are a new frontier in the field of image generation with artificial intelligence. These models overcome the limitations of traditional diffusion models by drastically reducing the sampling steps required to generate high-quality images. CMs are based on the concept of "Consistency Training" and the TrigFlow formulation, which simplifies calculations and increases stability during training. There are two main versions of CMs: discrete and continuous, each with specific advantages and disadvantages. CMs have shown significant performance in terms of image quality and computational efficiency, with promising results on the CIFAR-10 and ImageNet datasets. Future prospects include extending CMs to the generation of multimedia content like video and audio, accelerating the training of other artificial intelligence models, and integration with emerging technologies such as virtual reality and augmented reality. CMs are a technological evolution with significant implications for companies seeking to create high-quality visual content efficiently and sustainably.