Update information
V3.0 updated. Fine-tuned LoRA to improve the effects of generating characters with complex body limbs and backgrounds. The overall styling is more toward manga style rather than simple lineart. A lower CFG scale can make lines thinner.
V3.0更新。对LoRA进行了微调,改善生成人物肢体和复杂背景的效果。整体更趋向漫画风而非纯线稿。低CFG scale可以让线条变细。
Recommend settings
Model: Anything V4.5
VAE: Orangemix (the same with NAI)
LoRA Strength: 1
Sampler: DPM++ 2M Karras
Sampling steps: 20
CFG: 7
Negative embedding: EasyNegative、badhandv4
Highres fix is also recommended.
Note: all the LoRA names used in sample images are my local name, you need to change them to your saved LoRA filename!
基模:Anything V4.5
VAE:Orangemix(同NAI)
LoRA强度:1
采样器:DPM++ 2M Karras
采样步数:20
CFG:7
负面embedding:EasyNegative、badhandv4
推荐使用高清修复。
注意:样图中的LoRA名称是我本地文件名,要改成你保存的文件名!
Difference
Here's a comparision of V2.0 and V3.0 (locally called V3 and V4 on my machine - V2 on my machine is not published):
V2.0和V3.0的对比(在我的本地机器上命名为V3和V4——本地的V2我并未发布):
Comparision of different CFG scale:
不同CFG scale对比:
Description
FAQ
Comments (62)
请问大佬“Orangemix(同NAI)”指的是哪个啊,橘子模型太多了分不清楚啦(悲)
How to make this work I only get black images :(
Nice model!A good reference of sketch!
Would you please remove the trigger words in the next version?
All characters have the same face (
How to turn Safetensors into Lora? I am stuck :(
That's,Amazing
where do i find the vae organmix?
How do i find the vae organmix?
you can find them in the description of models, sometimes they have link to the various vaes like this page, at the top under the download button https://civitai.com/models/4437/abyssorangemix2-sfwsoft-nsfw
For everyone trying to figure out how to use this, on huggingface you find both the orangemix vae and the anything v4.5 pruned model that is used in the preview picture. just google them, cheers!
But how do you load the VAE in automatic1111?
@AbstractBunny562 you can, just go to settings and choose the VAE there!
请问怎么样图生图的时候,变成线稿?只能在文生图的时候变成黑白线稿,图生图模式炼不出来
我也想把自已是照片做成线稿,以便生成其他风格的AI图,但一直不成功,脸型都不是自已的,试过多个模型也一样。
我也是啊,用自己画的图生成线稿,一直空白要不就崩了
估计没办法彩色直接图生图,可以考虑先本地用PS把原图黑白化后,再试试,我先按这个思路走一下
你要转线稿用ControlNet的Lineart,用什么图生图....
图生图模式没法把彩图转成线稿,重绘幅度低了生出的图是和原图很像的彩图,重绘幅度高了是和原图毫无关系的线稿
+1
你要转线稿用ControlNet的Lineart,用什么图生图....
为什么我用这个lora生成的线稿背景都是灰色的?即使在prompt中加重white background也还是灰色背景
Hi, is it to make a version with for example trigger word for line thickness? I would be great to have control of the line to make them thinner sometimes. The results are great but I usually have drawing with very thick lines. Or maybe you can control the thickness of the line in another way?
RuntimeError: output with shape [320, 320, 1, 1] doesn't match the broadcast shape [320, 320, 320, 320] help me!!!!
this works fantastic with anything v3.
anything v4 and v4.5 has delete , have any model fit with anime lineart manga like this ? i use anything v3 but this worst , very bad quality , other model have a bad hand... :((( i need help
看到有人在说,要转线稿用ControlNet的Lineart,用什么图生图....这只是个lora,用来帮你稳定线稿画风的......
Two quick questions if you don't mind:
Can this be used with img2img? I haven't had any luck, and while I can get somewhat decent results using ControlNet Lineart in txt2img, I've seen a few very nice examples below that say they were made in img2img.
Also, has anyone had any success using this with other models besides Anything 4.5? Thank you very much.
May I know which artist is v3 based on?
Thank you for this!
Can't wait for the SDXL version :)
There used to be a video link here to a chinese site where you showed how to get lineart from any image. Why is it gone?
can you allow on site generation?
Edjjxjdjdjjjxc
Old school black and white manga style
We need a Pony version of this, Please
I am somehow using it with pony without it giving me a error message
@bigbeanaiart may you share the workflow please?
@krigeta I use forge not comfy unfortunately
This doesnt work in Forge :(
Really? Why?
@SafetyAction No clue but im onto bigger and better things hah
@yeenoghu Could you give me a hint? (working for another company? Playing with BETTER bigger models? studying AI, whatever? haha) I am curious. you can answer and let me read then delete if you want haha
@SafetyAction oh no, I actually started using it in the civitiai creator and got some good results and then started using the Krita interface and now im moving smoothly. Im sorry if i implied this model wasnt good, that wasnt my intention its actually been a big help to me.
@yeenoghu can you record your screen while you work on krita? I wanna see what people are doing with it to understand
@SafetyAction theres some really great tutorials by "Intelligent Image" that show everything on youtube.
@yeenoghu will check it out. Btw can I ask you kindly if you can share with me any character dataset (I want to do an experiment and see if everythign working fine with my computer, I wann abe sure the problme do not come from dataset. Any dataset of any character (not nsfw)) would be great, like anything can help, even if the lora exist. thanks
@SafetyAction I am not the guy to come to for that information.
@yeenoghu Arn't you a creator of loras? lol
@SafetyAction I made one, it sucked.
@yeenoghu oh my bad. I thought you were the maker of the lora:)
Hello, why does this appear on my research despite puting the SDXL filter? I dont undersatnd.
good
just adding the word (lineart) in positive prompts give better results than using this lora with all the anime models I tested. What a waste of time
The cost to train LoRA is less than the cost to train checkpoint. In addition, if you train a checkpoint overfitting to anime, it will not provide good results for other types like photorealistic. If you think this is a waste of time, try to train checkpoint or LoRA by yourself, instead of using other people's products.
d
Awesome!
Details
Files
animeoutlineV4_16.safetensors
Mirrors
animeLineartMangaLike_v30MangaLike.safetensors
16014_animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_161.safetensors
animeoutlineV4.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike (1).safetensors
线稿.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
5.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
nmlnrtstylv3.safetensors
animeLineartMangaLike_v30.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
LineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeoutlineV4_16.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
187_animeoutlineV4_16.safetensors
线稿_animeLineartMangaLike_v30MangaLike.safetensors
animeLineartMangaLike_v30MangaLike.safetensors
animeoutlineV4_16.safetensors
a4anime-lineart-manga-like-style.safetensors
animeoutlineV4_16.safetensors
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.















