2.5D Anime model based on AOM3 and Sakimichan style, realistic version is based on ChilloutMix, good at drawing anime characters that have a puffy look it come out with mature body, good with lora, good with NSFW, vae is baked in.
Positive Prompts:
(masterpiece), (best quality), HDR, intricate detail
For girls:
1girl, solo focus, skindentation, perfect face, puffy face, beautiful face, big eyes, puffy eyes, perfect eyes, eyelashes, __skin_color__, __eye_color, __hair__
Or
1girl, solo focus, skindentation, perfect face, puffy face, beautiful face, big eyes, puffy eyes, perfect eyes, eyelashes, __popular_character__
Use body parts for full body shot example: chubby body, narrow waist, big pelvis, wide hips
For Negative Prompts:
EasyNegative, bad-hands-5, (worst quality, low quality:1.4), low contrast, monochrome, text, watermark, thunder, skin imperfection
Make sure you have downloaded the embeddings (EasyNegative, bad-hands-5).
Use nsfw, naked, nude, nipples in the negative if you want clothes on your girls.
Recommanded settings:
Steps: 25
Sampler: DPM++ 2M Karras or UniPC
CFG scale: 7
Size: 512x768 or 512x512
Denoising strength: 0.45
Clip skip: 2
Hires upscale: 1.5
Hires upscaler: 4x-UltraSharp (can be found on huggingface.co)
Future improvement will be made to the model.
Known issues:
Third-party hands: I recommend cropping the photo using the Resize Width in the upscaler to avoid third-party hands.
Faraway eyes are not good.
Description
Trained with some outputs from v1.0, handpicked best pictures mostly NSFW.
FAQ
Comments (15)
There's a chance that the training is worsening the results of this model in comparison to the last version; wait to see, I guess, to compare the generations of both models.
I changed my monitor the old version didn't look as good on my new monitor then the older one, so I had to hand pick only good picture that looked best on the new monitor.
I felt this too from the images posted here.
it seems like images become very desaturated and blurry (like if VAE has not been loaded but less noticeable)
also face is not as 2.5d as previous version.
v1 and v2 are very different you still can use the old one if you like the old style.
actually the best model i ve used,
it can easily draw a thick and hot style.
heart for you~~
Amazing Model however it has difficulties with fully nude generations. It will often add some sort of ornamental necklace or sleeves. It's really really good, just needs some work generating bare necks/fully nude bodies.
Hope to see more content from you in the future. Love the model!
the images comes out blurry
Can you pls make the training data for v2 and v1 available?
Can you add the anime version to tensor.art?
dayum I'ma getting wet
This is an checkpoint right? Then. What is the base model for this one?
What a nice soft style! I really like this hazy anime look, It turns out very expressive and mysterious. Anatomy and diversity are also in order. Great model!