(250|50-10000) When creating an image, the denoising curve is subdivided into steps for processing. Each step (or iteration) involves the AI looking at subsets of the image called ‘cuts’ and calculating the ‘direction’ the image should be guided to be more like the prompt. Then it adjusts the image with the help of the diffusion denoiser, and moves to the next step.
Increasing steps will provide more opportunities for the AI to adjust the image, and each adjustment will be smaller, and thus will yield a more precise, detailed image. Increasing steps comes at the expense of longer render times. Also, while increasing steps should generally increase image quality, there is a diminishing return on additional steps beyond 250 - 500 steps. However, some intricate images can take 1000, 2000, or more steps. It is really up to the user.
Just know that the render time is directly related to the number of steps, and many other parameters have a major impact on image quality, without costing additional time.
Pages in category "Steps"
The following 6 pages are in this category, out of 6 total.