⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝🔥🐉 WND NOW UPDATED TO V2.0! 🐉🔥⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝⚝
FOR ALL INTENTS & PURPOSES THIS IS THE ANIME VERSION OF DNW!
It’s still a bit flavored to younger faces so (mature adult) may be necessary.
A bit less NSFW heavy but you can keep it in the negative to be sure.
Very variable and very good at fantasy! w/ amazing backgrounds!
***USE CLIP SKIP 2 FOR THIS MODEL! YOU CAN DO CLIP SKIP 1 AS WELL BUT 2 IS BEST!***
READ ME! THERE IS INFO ON HOW YOU CAN GET THE MOST OUT OF THIS MODEL!
There is a full & pruned version of this model available. This is a complex merge so even the pruned version is large as well.
There is no embedded VAE but it was designed to be used with this:
I recommend using the following negatives:
EasyNegative (The all around catch all. Always good in a bind)
badhandv4_anime (absolutely your go to neg for anime hands)
bad-hands-5 (good for the more realistic images but will not work amazingly for anime)
bad-artist \ bad-artist-anime (I’ve had good and bad luck)
bad_prompt \ bad_prompt_version2 (works very good sometimes)
(I typically use (bad-hands-5) & (bandhandv4_anime) at the same time to cover more bases)
*** Note: that the example images are provided as a basis of reference and raw output. In most cases untouched other than mild inpainting for the eyes or face in a few examples. *** THESE SHOULD BE EASILY REPRODUCIBLE RESULTS! IF YOU HAVE QUESTIONS PLEASE ASK 🗪
Note: There is a full & pruned version of this model available. This is a complex merge so even the pruned is also large as well.
Here are some optimal example settings for you to start with, but please keep in mind YMMV!
Steps 20-30 = Sweet Spot / 15-150
Sampler DPM++ 2M Karras, DPM++ SDE Karras, Euler A, DDIM,
CFG scale 4-9.5 = Sweet Spot / 3-20
Size 512x768, 768x512, 640x840, 840x640, 600x896, 896x600, (any size)
Clip skip 2
Hires Fix Enabled
Upscale by 1.5-2
Hires Steps Equal to Half or Exact amount of Steps or higher
Upscaler Remacri Upscaler, Lollypop Upscaler, 4x_NMKD_DigArt_10k|
_________________________________ (Older Info from version 1.5) _________________________________
A handpicked and curated merge of the best of the best in waifu’s, anime, 2D, and of course, fantasy environments, characters & creature designs. It doesn’t do everything but what it can do off the bat is a pretty damned amazing images in anime styles 2 - 2.5D quite effectively. 🎨
Specializes in illustrated, semi-realistic, painterly, subjective reality and a 2.5d aesthetic with semi real to real backgrounds. Backgrounds that look like high quality anime or 3d rendered, game backgrounds.
It does styles and illustration, and artists based prompts pretty damned well and adapts to most LORAS with ease, it DEFINITELY leans a bit towards nudity / NSFW and also more towards younger looking girls. (Look I didn’t plan this… the merges just made it that way.) I suggest a (((kids))), (((child))), (Young:1.5-2.0) in your negative and a (mature adult:1.5-2.0) in your positive.
However… it will also work with Clip Skip 1 better for prompts that are going after a specific art or artists styles. (Unsure if this is directly affected by LORA’s will have to test more) Example:
Notice that the art & artist style of the prompt comes out more in 1 and 2 has more of the industrial / cyber aspects. Play with it and see what works best for you.
SHOW ME WHAT YOU MAKE WITH THIS!! 😉
Here are some example image comparisons with some of the most common / top anime oriented models and also my previous model Dungeons N Waifu’s v2.2:
I was asked recently how to get higher quality images like the examples and about blurred eyes and faces:
Blurred eyes, or sometimes warped faces are not that uncommon in SD and sometimes require an inpaint to pull off. Most of my examples are not inpainted. But a few are.
First off when you run txt2img ALWAYS use HIRES FIX!
latent (nearest or exact) upscale / denoise 0.4-0.7 (Complete variable - depends on the type and complexity of the image. YMMV!)
Upscale by 1.5-2 but no more than that.
This will get you a pretty good base image to work with.
Afterwards… You can always send your txt2img picture to inpaint on the bottom right under the result area > select the face or eyes > lower the denoise strength to 0.5-0.57. > make sure your seed is set to random.
masked content = fill / inpaint area = whole picture ====== for faces
masked content = fill / inpaint area = only masked ====== for eyes
With “whole picture” selected for larger inpaint target aka faces (or hands) you may need the denoise to be lower to pick up more of the original image. 0.53-0.56
Now with “only masked” and a smaller target like eyes you get a little more freedom so your denoise can be 0.54-0.57
Run a batch of 4-5 (or more) for the faces… go through pick the best one… if you’re lucky you got the eyes and face in one go… if not you may need to choose the best face… send back to inpaint again… select just the eyes… and rerun a new batch (changing it to only masked and the settings described above for eyes.)
Finally when you get it all together send to extras:
Resize to 1.5-2
Upscaler 2 is what will get you there. If you want super sharp / realism images you choose NMKD
You want a good all around upscaler that can do nearly any subject you choose Remacri
If you’re working on toons, anime, animated, cell shaded you choose Lollypop
Each of these is unique and so is their outputs if you want to compare simply change the pull down and run off the various types and contrast / compare the finished results.
For the more advanced upscale: Get yourself Ultimate SD Upscale
(click the link for a basic tutorial on how to get it and how to use it.)