Seek Art license, Dreamlike license.
This model is something I stumbled upon while I was actually trying to make an anime model (still working on that....). The name came from the fact that I kept using the term 'That's just wild' for the outputs this produces. The colors are very vivid, with a lot of great details and fantastic, dramatic effects.
For best results use the Standard SD VAE.
Disclaimer: All prompts in the example images use InvokeAI Syntax NOT Automatic1111 syntax. The key thing you need to know is that 'term+' is equivalent to '(term)' ++ is the same as (()) and so on.
Version 2:
More realism. It's not quite perfectly photorealistic, but it can be pushed pretty close. Maybe even further than I've tried considering the amount of Protogen 5.3 that is in it. I really didn't expect to have anything to update here yet. I may have jumped the gun a little bit on releasing version 1 because it really wasn't up to where I wanted it, and it didn't match close enough to that original accidental find. It just didn't do justice to what I had seen and what I knew could be done. So I went back to that original accident and played around with it until I figured it out. But the eyes were still pretty bad, and hands were just absolutely monstrously terrible. I didn't think I was going to get anything going on this for a while, but then @duchaiten released version 1.2 of his AI Art mix, and it fixed all the issues I was having. So I get to provide an update after-all.
Recipe:
These were both add difference merges. To follow the instructions below, I'm using the basic formula in Automatic1111 Checkpoint merger:
Primary Model (A) + (Secondary Model(B) - Tertiary Model (C)) @ Multiplier (M)
Step 1:
WildMix_v1.1 = Protogen 5.3 + (DucHaitenAIArt 1.2 - SD base 1.5) @ 0.5
WildMix_v1.2 = WildMix_v1.1 + (Kenshi 1.0 - Anything-V3.0) @ 0.5
Promoted to V2
Version 1:
You will need the standard SD VAE for this one. It does also work with kl-f8-anime2.ckpt
This does include models that use Dreamlike, so here's the obligatory license.
Safetensor and ckpt files are available in both pruned and non-pruned versions. The pruned version is a fp16 no-ema.
It can do NSFW pretty well, but that's not really where it excels.
Models used in this mix:
Description
For this one I really wanted to figure out how to get some Synthwave Punk influence in there without making it too synthwave on every image. I think I achieved a pretty good balance with this version. While I was in there I also updated Duc Haiten AI art to version 2.0. (full recipe at the bottom). The result is better realism, still very saturated and vivid colors, and crazier/wilder hair and dramatic effects. It does have a tendency to put pink/blue/purple hair on the subject unless you tell it not to because of the synthwave punk influence. But it can do some pretty rad stuff because of that same influence. Example images use the same prompts as V2. Most of them use new seeds, but a couple I reused the seed just because the result was too cool not to. Also added a few new images to show off some of the new stuff it can do. trigger words are completely optional and the swpunk and synthwave trigger words can really change the style of the image. Some of the new examples are using the same prompts as others with just these trigger words added to highlight the differences.
Pruned versions are fp32 no-ema Prunes. It can be pruned to fp16 but be aware that the outputs will be different than those in the example images. Not terrible or worse quality, but they will be different.
Recipe:
WildMix_v2.1 = Protogen 5.3 + (DucHaitenAIArt 2.0 - SD base 1.5) @ 0.5
WildMix_v2.2 = WildMix_v2.1 + (Kenshi 1.0 - Anything-V3.0) @ 0.5
WildMix_v3 = WildMix_v2.2 + Synthpunksearch_raw @ 0.3