V2
===Instructions===
This model is merged for obtaining a specified style, this means during the merge process I keep using a specified promopt to test the affect, so I didn't test other prompts.
The outcomes are good. When the model uses style cues from sample images, it produces drawings that closely resemble finely-drawn hand-drawn lines. The style is similar to manga aesthetics.
Raw Version -> More hand-drawn style, less Digital Art feeling
V2a Version -> Good at drawing nude ,Colors is great! Digital Art feelings
V2b Version ->Between Raw and V2a Version
===Models Used For Merging===
XenotripAI - v1.0 | Stable Diffusion Checkpoint | Civitai
InkMix - Sensual1.0 | Stable Diffusion Checkpoint | Civitai
DogePoint - v1.0 | Stable Diffusion Checkpoint | Civitai
Galena Mix [Legacy] - Redux-v3.0 | Stable Diffusion Checkpoint | Civitai
AniToon - V1.0 | Stable Diffusion Checkpoint | Civitai
Crumpet Mix - 1.0 | Stable Diffusion Checkpoint | Civitai
---------V1----------
Old hand-drawn illustration style model.
If you are interested in the example images, you can use the following prompt format:
==Prompt==
Raw,illustration with brushstrokes,official art,(hatching \(texture\)),(muted color,partially colored:0.8),detailed hatching,(drawing_trace),flat color,1990s \(style\),recolored,
neon genesis evangelion,
<Fill in the content you want here, It is recommended to use natural language to describe, as using tags can cause some problems (SEE BELOW)>,
art by sadamoto yoshiyuki
==Negative prompt:==
(worst quality, low quality:1.4),(bad anatomy),(bad hands),digital art,1980s \(style\),twisted lines,chaotic lines,over exposure,drawing
==Other settings==
Steps: 30, Sampler: DDIM, CFG scale: 7, Seed: 2326670396, Size: 768x768 ENSD: 31337
vae: sd 86000 clip:1 or 2
You can also try some negetive embbeding like EasyNegetive, bad-artist and bad-image.
-->Known issue:
If you are using the prompt above, you may find that the character sometimes is holding a pencil/sketch/paper and painting tools often appear in the image, this is caused by tag I used and I haven't found a way to fix this problem yet.
Description
===Instructions===
This model is merged for obtaining a specified style, this means during the merge process I keep using a specified promopt to test the affect, so I didn't test other prompts.
The outcomes are good. When the model uses style cues from sample images, it produces drawings that closely resemble finely-drawn hand-drawn lines. The style is similar to anime aesthetics.
Raw Version -> More hand-drawn style, less Digital Art feeling
V2a Version -> Good at drawing nude ,Colors is great! Digital Art feelings
V2b Version ->Between Raw and V2a Version
===Models Used For Merging===
XenotripAI - v1.0 | Stable Diffusion Checkpoint | Civitai
InkMix - Sensual1.0 | Stable Diffusion Checkpoint | Civitai
DogePoint - v1.0 | Stable Diffusion Checkpoint | Civitai
Galena Mix [Legacy] - Redux-v3.0 | Stable Diffusion Checkpoint | Civitai
AniToon - V1.0 | Stable Diffusion Checkpoint | Civitai
Crumpet Mix - 1.0 | Stable Diffusion Checkpoint | Civitai
FAQ
Comments (13)
What an incredible and responsive model. VERY GOOD WORK! And I see what you mean by being narrative instead of using keywords, I never thought to try that with any model before, but it works so damn well. Are a lot of models more responsive to narrative prompts now?
I think sd1.5 can response to narrative prompt.well, so the models based on sd1.5.can response.well.But I.just want to fix the problem caused by illustration/hatching tagsXD.
ANyway thanks for ur nice comment!
(worst quality, low quality)
The choice of keywords significantly influences the overall impression of the picture.
Can the next version reduce the reliance on negative keywords
Cant do that, cause this is a model based on merging other than training.
Also my purpose for making this model is creating specific art style, so I have no passion to test other art style prompt, sorry.
Anyway, Thanks for ur comment and ur post! Thank u for testing my model. I'm not a good model maker, so I just release this model when it achieved my desire art style.
@noesca You've done a very good job, and the quality of your model is very high!
@li1336797007741 thank u! love u!
Hey!
I tried to replicate this effect on my Lora, Hand Draw Manga | Style, for Illustrious, it's close, but not as perfect as yours!
Amazing job!
Thank u very much!!!
Glad to see ur lora! Honestly I cant create those images on comfyUI when i tried use the same model like I used to do in the past, maybe it's because these images were created on auto1111 webui :D
@noesca Oh, yeah, i could not get the same images too, but, i manage get some nice gens either way xD
I used 2 ksamples (1 to generate the image, upscale, then 1 to hiresfix), DPM++ 2M SDE Normal, CFG scale: 7.0 on both, 28 (or less) and 28 steps, 0.46 denoise on the 2th ksampler, and some negative embeddings
This worked really nice, my pc is kinda slow, but i would do some handfix too if i could xD
@noesca Oh! that checkpoint seens really interesting indeed!
I using this one quite o lot locally: https://civitai.com/models/1673823?modelVersionId=2560398