Rainbowpatch is a fine-tuned model specifically trained to generate high-quality stylized images, this model doesn't require complicated detailed text prompts. It is capable of generating high-resolution images with only a few words or a simple description of the subject. If you get bad image quality then increase the sample steps to generate high-resolution images. Rainbowpatch is less dependent on negative prompts. you can generate good images without any negative prompts if you're generating a simple illustration, I didn't use any negative prompts to generate all the sample images I uploaded here. Use a negative prompt if you get any unexpected results. Here are some prompts examplesPrompt: Portrait of a beautiful Woman Negative prompts: non Prompt: Portrait of a beautiful superhero woman Negative prompts: non Prompt: A beautiful Portrait of Wonder woman, a detailed face, (detailed eyes), (rainbowpatch:0.5) Negative prompt: Logo, watermark, double faces, text, Deformed, blurry, ((bad anatomy)), disfigured, poorly drawn face, mutation, mutated, extra limb, ugly, poorly drawn hands, missing limb, blurry, floating limbs, disconnected limbs, malformed hands, blur, out of focus, long neck, long body, (((mutated hands and fingers))), (((out of frame))) Steps: 30, Sampler: Euler a, CFG scale: 7, Seed: 1670235636, Size: 512x768, Model hash: d9d11a9ff2 You can use the trigger word (rainbowpatch) if you want. normally it doesn't require you to use that because the style is heavily implemented in the model. This is a retrained version of the original Rainbowpatch model. |