Downloads · 30 days
6
1% of all-time downloads
shirome
ottopilot/shirome
shirome is a text-to-image model from ottopilot. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as cc-by-nc-4.0.
All-time downloads
508
Public
Repo size
43.6 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors37.9 MB · 87%
At a glance
- Task
- Text-to-Image
- Library
- diffusers
- License
- cc-by-nc-4.0
- Access
- Public
- Created
- Feb 3, 2024
- Updated
- Feb 3, 2024
- SHA
- 7c846f0f
Try a prompt
Base models
Model card
From the Hugging Face model README
Shirome (whitewashed eyes)
<Gallery />Model description
Now you too can reduce your waifus to blissfully mindless drones! Shout out to subliminalbo on Tumblr/DeviantArt for sources and support.
Trigger word is "shirome" because "whitewashed eyes" is actually biased towards white hair and blue eyes, which really explains some of my previous results. Also, works with men so you can make your own himbo army.
This is still finicky, hence the versioning. It has trouble producing consistent results sometimes (pupils still visible or not matching the sclera). Tips for use:
- negative prompt "pupils"
- hires fix recommended
- don't use Adetailer mediapipe_face_mesh_eyes_only, it's putting detail back in the eyes - detail you actually don't want if you want a nice, smooth, pupilless eye.
Trigger words
You should use shirome to trigger the image generation.
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.
- Task
- Text-to-Image
- Library
- diffusers
- License
- cc-by-nc-4.0
- Created
- Feb 3, 2024
- Updated
- Feb 3, 2024