Welcome to MirrorMask - This is a fine-tuned Stable Diffusion model trained from the 2005 dark fantasy film MirrorMask.
Use the token mirrormask in your prompts for the style to show. Examples down below:
An example of calling mirrormask before the prompt (mirrormask a woman)
Single SamplesAn example of calling mirrormask with a comma before the prompt (mirrormask, a woman)
Single SamplesAn example of calling mirrormask with the additional word of "by" before the prompt (mirrormask by a woman)
Single SamplesAn example of calling mirrormask with the additional word of "in" before the prompt (mirrormask in a woman)
Single SamplesAn example of calling the prompt then mirrormask (a woman mirrormask)
Single SamplesAn example of calling the prompt then adding a comma before mirrormask (a woman, mirrormask)
Single SamplesAn example of calling the prompt then adding the word "in" before mirrormask (a woman in mirrormask)
Single SamplesAn example of calling the prompt then adding the word "by" before mirrormask (a woman by mirrormask)
Single SamplesFor more variety try adding the word "style" after the token word mirrormask (no examples given to save space).
This model can be used just as you would any other Stable Diffusion style model. For more information,
please have a look at the
Stable Diffusion.
You can also export the model to
ONNX,
MPS and/or
FLAX/JAX.
1from diffusers import StableDiffusionPipeline
2import torch
3model_id = "GeneralAwareness/MirrorMask"
4pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
5pipe = pipe.to("cuda")
6prompt = "a woman by mirrormask"
7image = pipe(prompt).images[0]
8image.save("./awoman.png")
This model is under a creative commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
To see what rights you have under this licence follow this link -
https://creativecommons.org/licenses/by-nc-sa/4.0/
