Merged model for anime style.
No VAE.
All image are generated in Stable Diffusion A1111 web-ui at 768x1024, 18 step, 11cfg, Sampler: DPM++ 2M Karras.
No face restore.
All sample image have prompt attached.
Suggested negative: (worst quality, low quality:1.5), bad anatomy, fewer digits, text, old, signature, watermark, username, artist name, bad proportions
Model made for simple and advance prompt.
Check my other models.
You can marge this model with other model, but giving back-link to this model on your description.
About model:
This model is intended for free use and free sharing image.
Not use it for selling generated image or merged checkpoint.
This is long time merging, testing, merging, etc... so don't ask for details
because i don't remember all of it( sorry).
About you:
Feel free to test and give feedback, detailed feedback get me know what work or not work, so i do my better to fix on next release.
Description
Initial release.
More feedback, more quality on next release.
FAQ
Comments (6)
This looks amazing, thank you! Would you mind also uploading the larger unpruned version if you have one?
Thank you. Sorry, i don't have unpruned models.
Seems to work better with NAI vae than without. Cool model
>don't ask for details
>because i don't remember all of it
Then you do remember some details, so I'm gonna ask anyway, can you give some examples of some of the checkpoints that went into this merge? It has big implications on how it's best prompted
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.