“seed”: 45543679,

“prompt”: "(((full body))), portrait of cute redhead girl, (((on her knees))), (((legs apart))), (((nude))), ((very very slim)), (small breasts), (smooth pussy), wispy hair, realism, high-quality rendering, stunning art, high quality, film grain, Fujifilm XT3, dreamy, freckled, blemish, 85mm, nikkor, fuji provia, fuji portra, night, backlight, ",

“negative_prompt”: “wide angle, NG_DeepNagetive_V1_75T, (greyscale:1.2), (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), day, sun, sunlight”,

“cfg_scale”: 5,

“width”: 512,

“height”: 768

    • Crunchypotat77@vlemmy.netOP
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 year ago

      It’s an alternate syntax to add weight.

      From the website:

      “The prompt tells the AI what to generate. 1. order matters: text in the beginning are treated more seriously by AI. 2. emphasis: (red dress) -> red dress is 10% more important to AI; ((red dress)) -> another 10%; (red dress:1.4)->40% more important. Do not go over 40 as it will likely turn weird.”