If you want to use more my checkpoint online generation, please visit here.
https://tensor.art/u/762555264535746522
You can run WAI-ANI-NSFW-PONYXL and use its API on SinkIn:
V14 released.
Increased composition diversity and improved model accuracy now allow images to be generated with fewer steps.
Recommended Setting
Steps:20-30
CFG scale: 5-7
Sampler: Euler a/DPM++ 2M Karras
ADetailer face_yolov8n/s.pt use can fix eyes
For the example images
I used 1024x1360,
generated directly without AD or hires fix.
Positive Prompt
score_9, score_8_up, score_7_up,source_anime,Negative Prompt
worst quality,bad quality,jpeg artifacts, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark,V13 released.
Increased body stability and accuracy.
overall balance adjustment.
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a/DPM++ 2M Karras
ADetailer face_yolov8n/s.pt use can fix eyes
For the example images
I used 1024x1360,
generated directly without AD or hires fix.
Positive Prompt
score_9, score_8_up, score_7_up,source_anime,Negative Prompt
worst quality,bad quality,jpeg artifacts, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark,V12 released.
Style adjustment, background adjustment, and increased stability.
Added more compositions.
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a/DPM++ 2M Karras
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt
score_9, score_8_up, score_7_up,source_anime,Negative Prompt
score_6, score_5, score_4, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark,v11
Weight adjustment, balancing the art style
All example images(Except the cover) are generated at 1024x1360, without using AD and hires fix.
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a/DPM++ 2M Karras
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt
score_9, score_8_up, score_7_up,source_anime,Negative Prompt
worst quality,bad quality,jpeg artifacts, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark, V10
Better Background
Clothes details up
Better mature female
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a/DPM++ 2M Karras
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt
score_9, score_8_up, score_7_up,source_anime,Negative Prompt
worst quality,bad quality,jpeg artifacts, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark, ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
V9 Hyper12step
Please refer to the instructions on the right side of V9 Hyper12step for usage.
V9 released.
v8->v9
adjusted the facial data to make it more versatile, allowing for easier generation of faces across various ages.
added more NSFW materials.
attempted to reduce the generation of some unnecessary details.
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a/DPM++ 2M Karras
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt
score_9, score_8_up, score_7_up,source_anime,Negative Prompt
worst quality,bad quality,jpeg artifacts, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark, __________________________________________________________________________________________________
hyper12step released.
about size
Various combinations from 768x768 to 1360x1360 can produce good images.
Recommended Setting
Steps: 12
CFG scale: 3.5
Sampler: Euler a
Not very necessary to use ADetailer correction
Positive Prompt
score_9, score_8_up, score_7_up,source_anime,Negative Prompt
score_6, score_5, score_4, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark,PS.The example images use a size of 896x1192. no AD no Hires. fix
__________________________________________________________________________________________________
Version 8 released.
v7->v8
Added some training related to camera angles and backgrounds.
Added more NSFW materials for additional training.
Added some training for special effects.
Optimization of overall composition logic.
Fixed some minor issues.
better eyes and hands.
Recommended Setting
Same as v7
__________________________________________________________________________________________________
Version 7
v6->v7
Fixed the issue where images appeared entirely covered with snowflakes.
Revised the composition logic to be more suitable for general scenarios.
Added new assets for backgrounds, mature women, and clothing.
Ensured consistency with the previous model and further fixed the issue of blurry eyes in medium and long shots.
Fixed some minor issues.
V7 Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a/DPM++ 2M Karras
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt
score_9, score_8_up, score_7_up,source_anime,Negative Prompt
score_6, score_5, score_4, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark,________________________________________________________________________________________________
Regarding the 8step model, it is a lightweight model that can generate images in as few as 8 steps. However, the quality and text comprehension will be relatively lower. If speed is not a major concern for you, it is recommended to continue using version 6.0.
8 STEP Setting
Steps: 8-12
CFG scale: 3
Sampler: Euler a
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt and Negative Prompt same 6.0
__________________________________________________________________________________________________
6.0
v5->v6 Change point
Better Background
better eyes
fix vae
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a/DPM++ 2M Karras(better eyes)
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt
score_9, score_8_up, score_7_up,source_anime,BREAKNegative Prompt
score_6, score_5, score_4, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark,You can view the Supported art styles from https://civarchive.com/articles/5715
Thank you, DeepDark_Fantasy514, for providing the test.
v4->v5 Change point
Better Background
Clothes details up
Add some mature female
Little better hands and eyes
Recommended Setting
Steps: 30
CFG scale: 7
Sampler: Euler a
ADetailer face_yolov8n/s.pt use can fix eyes
Positive Prompt
score_9, score_8_up, score_7_up,Negative Prompt (THX for snipsnapsnipsnap recommend )
Base:score_6, score_5, score_4, source_cartoon
Optional: source_furry, source_pony can bleaches skin
else: 3d, (censor),monochrome,blurry, lowres,watermark,
I often use it now
[score_6, score_5, score_4, source_cartoon,
3d, (censor),monochrome,blurry, lowres,watermark, ]
Description
FAQ
Comments (42)
WAI-NSFW-illustrious-SDXL是不是比这个更适合动漫啊
你指的动漫是什么，动画角色吗，那确实光辉更合适
will there be a 12 step finetune for every new version or were 8/9 just one-offs?
感谢作者大大一直更新！不过我个人感觉V9比V10要更好点，因为在同样的参数和正负提示词的情况下，V10极大概率出现半写实的画风（最明显的地方就在于人物的鼻子会出现两个鼻孔，面部总体也更偏3D画风了），而V9则会正确地生成动漫画风。
我把source_cartoon，3d，realistic都加进了负面提示词还是没法覆盖掉半写实画风。请问是因为训练新版本模型时增加了半写实的权重吗？
My new favorite checkpoint model (v10). Keep up the great work!
I feel like the V10 is weighting a bit too much on mature female? I mean I use the same exact prompt as the V8 and V9 and the V10 is looking a bit more mature than the others. Well, it's not bad as I can just use age slider Lora but just a head-up for anyone using preset prompts and finds the result is older.
It's beautiful! <3
New version forces too much realistic style and lose anime... Not good
Plus, can't use v8 or v9 on site anymore...
@CathyPowers About style , I made significant adjustments to the weights this time, so there may be considerable deviations. If many people don't like it, I will revert to the original style in the next version, so please wait for the next update. As for online generation, since only two online generations can exist at the same time, only the latest version can be used
@WAI0731 what's the other version/model you have online?
@WAI0731 ah, used it today, not good with more than one character... Btw you change a lot huh? I remember that V7 wai are amazing, than change to V8 and now V9 really fast
@CathyPowers My update speed is not very fast, basically one update per month.
@WAI0731 couldn't you merge all versions in one with a selectable option? Like https://civitai.com/models/264290?modelVersionId=882225
@CathyPowers Editing is too labor-intensive, and the documentation is hard to write. Plus, I have to re-upload the models, which are several hundred gigabytes.
@WAI0731 I see, so it's possible but it's hard to do, okay
@CathyPowers Each model has different recommended settings, and when you have dozens of models grouped together, not only is the documentation hard to write, but it’s also difficult for others to understand.
@WAI0731 okay
kinda liked V9 more, could have left it available.
Sorry, because I need to gather feedback on V10 to facilitate the updates for the next version.
@WAI0731 Is it possible to left v9 available, bec my lora work so perfectly with ur checkpoint, i'd tried other checkpoint but non of them is as good as urs
@Meowpass Each person on Civitai can only have 2 online generations. If I give V9, the online generation for my other model will be disabled
@WAI0731 Ahh I see
But V9 is available? https://civitai.com/models/404154?modelVersionId=931577
感谢大佬更新！ 服装细节与背景细节有很大改善，但人物面部变得偏写实，希望可以回到像V9，V8的风格。
在v10版本应用原v9几乎不出坏图的预设，结果有小一半鬼图，其余的也不尽如人意。
是否可以重新开放v9的在线生成权限呢？🤗
A lot of artifacting in v10 compared to v8 and v9 with same settings, but I do like the background detail. Going back to 9 for now. It's pretty much perfect.
Hopefully V11 will combine the best parts of this version and the previous ones.
Has a really hard time doing dark skin on women, face and breasts in particular.
pretty hard make view from behind without face of char.
using this vae (vae-ft-mse-840000-ema-pruned) seems to fix the realism issue for me.
I really liked the 10th version, but I’m forced to switch back to the 9th version for several reasons. With many LORE the characters faces look too much realistic, which doesn’t fit the concept of the previous versions. The style of the 10th version differs too much from the anime aesthetic of the 9th and 8th versions. After conducting several tests, I decided to return to the 9th version until the style of the 10th or even the upcoming 11th version aligns more with the overall concept. Thank you.
Don't agree, this makes the model much more versatile and all you need to do in order to get anime style characters is say anime style in the prompt.
@apefordead568 My opinion based on the tests. The “source_anime” prompt doesn't always work.
It's beautiful and I can't stop!!
EDIT: V11 is out to hopefully fix what I observed here, can't wait to try!
I feel the style of V10 has shifted too much. One reason I love this model and keep using older versions is its colorful palette and how it stayed "firmly" 2D. Moving towards the 2.5D of too many other models would make it just one more of them than a special one. Hopefully next version will go "back to the roots" soon!
Details
Available On (28 platforms)
Same model published on other platforms. May have additional downloads or version variants.