HOME | DD

Description
I noticed that some people had mastered Stable Distortion to such a degree that they could create masterful artwork. I asked Pandorasaurus to make a picture of Asuna, the Planar of the Moon, and ended up with FIFTEEN different pictures. Some were too close to the original, while some were great but had outfits that did not fit.

This is probably the one that is the closest in terms of emotion and physique to Asuna, even if the outfit is different.

Here is the total input required to pull this off, quoted directly from Pandorasaurus:

"I used Img2Img with your image, and created several images at 0.4, 0.6, and 0.8 prompt strength.
girl with white hair and intricate jewelry, fantasy, by Ilya Kuvshinov, by WLOP, by Artgerm, by Makoto Shinkai, highly detailed, masterpiece, best quality, highres, absurdres, beautiful face, beautiful body, gorgeous, deep eyes, realistic shaded, stunning, intricate, hyper-detailed, sharp, anime, manga

Width: 512
Height: 768
Steps: 45
Guidance Scale: 8.0
Use Face Correction: None
Use Upscaling: RealESRGAN_x4plus
Sampler: ddim
Negative Prompt: ugly, fat, obesity, chubby, deformed, deformity, ugliness, blurry, disfigured, poorly drawn face, mutation, mutated, extra_limb, extra limbs, wrong hands, poorly drawn hands, messy drawing, text, cropped head, bad anatomy, two heads, child, kid, gross, mutilated, disgusting, horrible, scary, evil, old, conjoined, lowres, extra digit, fewer digits, worst quality, low quality, jpeg artifacts, watermark, username, artist name, censored
Stable Diffusion Model: Anything-V3.0-pruned.ckpt"

Description
I noticed that some people had mastered Stable Distortion to such a degree that they could create masterful artwork. I asked Pandorasaurus to make a picture of Asuna, the Planar of the Moon, and ended up with FIFTEEN different pictures. Some were too close to the original, while some were great but had outfits that did not fit.

This is probably the one that is the closest in terms of emotion and physique to Asuna, even if the outfit is different.

Here is the total input required to pull this off, quoted directly from Pandorasaurus:

"I used Img2Img with your image, and created several images at 0.4, 0.6, and 0.8 prompt strength.
girl with white hair and intricate jewelry, fantasy, by Ilya Kuvshinov, by WLOP, by Artgerm, by Makoto Shinkai, highly detailed, masterpiece, best quality, highres, absurdres, beautiful face, beautiful body, gorgeous, deep eyes, realistic shaded, stunning, intricate, hyper-detailed, sharp, anime, manga

Width: 512
Height: 768
Steps: 45
Guidance Scale: 8.0
Use Face Correction: None
Use Upscaling: RealESRGAN_x4plus
Sampler: ddim
Negative Prompt: ugly, fat, obesity, chubby, deformed, deformity, ugliness, blurry, disfigured, poorly drawn face, mutation, mutated, extra_limb, extra limbs, wrong hands, poorly drawn hands, messy drawing, text, cropped head, bad anatomy, two heads, child, kid, gross, mutilated, disgusting, horrible, scary, evil, old, conjoined, lowres, extra digit, fewer digits, worst quality, low quality, jpeg artifacts, watermark, username, artist name, censored
Stable Diffusion Model: Anything-V3.0-pruned.ckpt"