Idea: AI generator/Text to image: Rather than generating 3 images using same model/sampler, allow a user to generate X images from the same prompt, but using a selection of X models or X samplers (advanced mode) to do so. Would reduce the trial and error of seeing which models respond best to which prompts?