About V2-
This version was not made on a base of 1.5 but with a mix of models, i don't remember which as it was made early this year just after the previous one. It should be more "useful" as it can do both "my style" using -vktr- trigger (try weights 0.5 up to 1.5 or more for full glitch) and without the trigger it acts as a normal model and is pretty versatile. Hope you'll like and share your results✌️
Edit: Very good results with the addition of Detail Slider - LoRA weight 2
Edit 2 : I do not consider V2 superior to V1, they are simply different. Try both if you wish.
Hey
This model is a bit special.
It was made earlier this year, and i've been using it to make art that you can see by clicking on the Pinterest link on my profile.
Unfortunately i can't use SD on my computer anymore, and the easiest way for me now is to generate through Civitai.
The thing about this model is that it's a base of SD 1.5 trained on my own art, that i've been making since 2016.
It has a very strong aesthetic and the model is mostly used as a conceptual tool.
Elements and references that composes it includes: flat black/white/red (don't hesitate to force the red color by adding it in your prompt), inked style, heavy textures, distortion, grain, bloom, a hint of eastern european graphic art, 2 tbsp of Megastructures, female androids, 19th century German Romanticism etc…
-
It'd be nice to share your creations on here. Peace
Description
FAQ
Comments (13)
great work and refreshing aesthetic! can we have a safetensor version please? would love to try it out.
Really cool! Nice style
you,my friend
you are the TURE HERO!
wow, this is awesome! are you ever going to make an SDXL version?
Love this! Assumed you were a Nihei fan then saw your LoRAs - going to check those out too, nice one :)
love this style and the model<3 thank you for making it!!
This is so cool that I don't even know what to write!
Love the art style. I found hiding pictures within the generated image to be super fun
I really loved this art style. thanks so much for this... I just want to know more about this style. What is this style called?
safetensor version?
why?
raise NotImplementedError(error_message)
NotImplementedError: You have selected base model other than SDXL. This is not supported yet.
God I love dicking around with this checkpoint but mostly I don't ever feel like I'm using it right lol.
It's very, very different and I appreciate that. The future of Stable Diffusion IMO is going to be unique checkpoints to suit certain tasks as opposed to mega-checkpoints that are supposed to do everything. And this is exactly what I mean by that.
V1. Unfortunately, my testing text prompts turned out to be completely incompatible with this model and the pictures came out just defective. I want to say that this is a really specific stylistic model, and it is not worth approaching it with the usual prompts and LORAs. Nevertheless, from what I saw in the examples of the author's works, it is really very original and unusual. And I want to express special respect for the fact that he created it based on his OWN works! To me, it looks real like science fiction, when an artist teaches artificial intelligence to draw like him. This is very exciting for the imagination (considering that I am also a little an artist by education and work)!
V2. It looks more comfortable and obedient, very flexible, diverse, and the pictures turn out much more correctly, with an interesting muted color style. But, unfortunately, it is very unstable and unpredictable in results.
Nevertheless, thank you for your work! If you can make it more versatile and stable, it will be just a bomb on CivitAI! :))
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.