metadata

tags:
  - text-to-image
  - stable-diffusion
  - lora
  - diffusers
  - template:sd-lora
widget:
  - text: >-
      masterpiece, high quality, defined pupil, looking at viewer, rounded
      pupil, defined iris, (soft iris:1.2),
    parameters:
      negative_prompt: >-
        bad_anatomy, deformation, amputation, deformity, deformed_nipples,
        duplicated_torso,  deformed_torso, long_torso, large_torso,
        unproportioned_torso, (deformed_pussy:1.2), (deformed_hands:1.2), 
        unproportioned_eyes, unproportioned_head, small_head, duplicated_nose,
        big_nose, fusioned_clothes, fusioned_arms, undefined_limbs,
        divided_pussy, red_pussy, duplicated_pussy,  deformed_anus,
        deformed_pussy,
    output:
      url: images/umi_final.png
base_model: cagliostrolab/animagine-xl-3.0
instance_prompt: id_umi_sonoda
license: mit

Umi Sonoda

Prompt
masterpiece, high quality, defined pupil, looking at viewer, rounded pupil, defined iris, (soft iris:1.2),

Negative Prompt
bad_anatomy, deformation, amputation, deformity, deformed_nipples, duplicated_torso, deformed_torso, long_torso, large_torso, unproportioned_torso, (deformed_pussy:1.2), (deformed_hands:1.2), unproportioned_eyes, unproportioned_head, small_head, duplicated_nose, big_nose, fusioned_clothes, fusioned_arms, undefined_limbs, divided_pussy, red_pussy, duplicated_pussy, deformed_anus, deformed_pussy,

Model description

This model was trained to generate high quality images based on SIFAS cards.

To achieve better quality, you should be using hako-mikan's regional prompter, along with Latent Mode, which modifies the way Stable Diffusion isolates the LoRA resulting in a significant improvement.

Trigger words

You should use id_umi_sonoda to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.