• Xyre@lemmus.org
    link
    fedilink
    English
    arrow-up
    4
    ·
    1 year ago

    Is it drawing characters by default or are you providing additional context in the prompt?

    • Krank Star@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      8
      ·
      1 year ago

      no context only input is the username

      1000011819

      Xyre Steps: 20, Sampler: DPM++ 3M SDE Karras, CFG scale: 7, Seed: 189656042, Size: 768x1280, Model hash: 74dda471cc, Model: realvisxlV20_v20Bakedvae, Clip skip: 2, RNG: CPU, Version: v1.6.0

    • pewter@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      ·
      1 year ago

      Depends on what the model was trained on. A good way to test is to use a prompt that should give you nothing. I just tried stable diffusion with the prompt “.” and got an overhead shot of a rustic dinner table. I tried the dreamshaper and it gave me a headshot of a woman.

    • tal@lemmy.today
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 year ago

      Xyre

      Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 0, Size: 1024x1024, Model hash: ebf42d1fae, Model: realmixXL_v15, Token merging ratio: 0.5, Version: v1.5.1