Prompt Example: Wetsuits

Now that the Silverstar model is published, would like to provide some examples of prompting. The first, by popular demand, is wetsuits!

We shall name him "Simon"

Positive Prompt:

(1man), photo of (young male, {21|22|23}yo, {slim|skinny|svelte|thin|fit}, {cute|handsome|attractive|average|}:1.3), on a beach, wearing (skintight highly detailed {black|charcoal|camo|navy} and {black|white|red|blue|teal|cyan|lime|green|orange|yellow|lime|silver} {neoprene|rubber|spandex|pvc} wetsuit), ({cute|edgy|short|medium length|sidecut|parted|straight} hair:1.2), ({blonde haired male|ginger male|black haired|auburn haired|}:1.3), slim thighs, ({morning|noon|afternoon|evening|night}, {sunny|backlit|sunset}:1.5), candid, breathing, head tilt, {seductive smile|smirk|smile|grin|}, expressive, from side, (tack focus, absurdres)

That is a bit, first lets cover those items in the brackets, ie {21|22|23}. I use dynamic prompts extension in A1111. This causes a random selection of the items in seperated by the pipe (|). So for the age, each generation in a batch will select one of 21, 22, or 23. This allows getting quite a bit of variety each batch. If you don't use A1111 or can't use the extension, just convert those to your desired keyword and generate.

Negative Prompt:

(muscles, pecs, abs, nipples, zipper), (Low quality:1.5), boring, grotesque, distorted, pixelated, zombie, bad hands, deformed, 3D, Sketch, Bad Anatomy, bad art, bad artist, (deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime:1.4), text, close up, cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, fat, obese

These were generated at 512x768, Using Euler A, 20 Steps, CFG 7. Increasing steps will increase quality/texture somewhat. Even higher quality/resolution? I personally like to send my results to img2img at 768x1152 res. Use a denoise of .35 if you mainly want the same image but increase quality/resolution, or if you want variety use .75. Be mindful of the prompt in img2img, with dynamic prompts and high denoise you can get quite a bit of variety.