Massive thanks to @NicoGJ and @BetterThanNothing for their amazing models Toonify and Prancing Invasion Ribbon Wave.
ABOUT V2 (more idiocy)
Honestly i was a bit dissapointed with V1, the idiotic version. i mainly upload these models for myself, and out of spite for how generic looking and annoying to prompt a lot of other models are. the idiotic version felt weird to prompt though, and looking at the prompts used for the showcase images, you can see that they are longer on average than my other models. anatomy too wasnt the best. fingers and eyes were wonky at times, so i created more idiocy to fix these issues!! lets get into it
WHAT DID MORE IDIOCY FIX??!??!?!
-Prompting is WAYYYYYYY better. its definetily as simple and fun to use as my other models now, with the last version i felt that i needed overprompt a bit more than my other models, but this is fixed. prompt following, ease of use, and fun factor is up to standard!
-Anatomy is better. eyes specifically were really inconsistent in the last version, but this is better now, and hands got a really nice upgrade as well
This Checkpoint took forever to make. my last checkpoint, the stupid mix, was a more adult web comic style, and the dumb mix before that satisifed a more rule34 almost internet drawing vibe. i didnt know where to go for this one but i settled on something a bit closer to a classic anime style, while still giving it a unique identity. i was super picky about this because the whole goal for these models was based on my hatred for the millions of average, basic styled models. but i think i managed to avoid that disgusting mediocrity and boring style, while also cadering towards the more modern anime style!
This model is one i can imagine will be trained on and manipulated most by the community, it offers a clean style that will adapt to most loras and manipulation you throw at it, and it also has my signature feel. as with all of my models, this one is just as easy to work with, im never gonna release a model that has dumb prompting guidelines. you can have fun with this one!
Strengths (more ego inflation)
-Compatibility. this model is closer to the more modern styles of anime, so most loras, controlnets, and training will work with this one.
-Anatomy. anatomy is pretty great with this one. hands and eyes will not be a problem here
-Pony Score Tags, i have never even used these in any image ive ever made with this model, maybe its better, maybe not, but they are certainly not needed
-Overall its really just well rounded. it does about everything you would want pretty well. its my least experimental model, but it has really good character design, nsfw, backgrounds, prompting, etc. theres no big weakness i can think of.
Weaknesses (trying not to be biased ðŸ˜)
-A small bit of bleeding. this means that the model has a slight tendancy to put sleeves/a jacket on your characters. its just due to the nature of the high level of merging i've done here. this is very easily fixed and will be barely noticable if your prompting for any sort of clothes (or naked), which most people do to some extent, but i figured it was worth mentioning
Settings. All of the showcase images you see are completely stock with no post editing, no fancy comfyUI nodes, no controlnet, no inpainting. the only "post" thing i do is a latent upscale (hi res fix in forge and A1111). ive used ComfyUI (stock nodes), and Forge for the images seen, and copying the image into comfy/forge should give you the exact shit i did. i have also used the model in Fooocus, as well as sites like SeaArt, with no problems.
also, most images were first or second rolls, and i always gen 2 images at a time, generally all of the images shown were made in around 4-8 total images
for transparency sake i figured i would mention these. i can be confident in saying you wont need to reroll much due to awful anatomy, incoherent images, or artifacts
Steps: 30 or more, never tested under that, but im considering a lightning version of this. lemme know if u want that
Sampler: any of the DPM's with karras lol, im sure others will work fine
Negative Prompt: just used a copy paste i made years ago, as seen in the preview images, big negatives wont effect quality negatively, and small ones proly wont decrease quality much idk, just put watermark tags in there
Latent Upscale For ComfyUI: i use bislerp at the same Ksampler settings as your base image, and a denoise of 0.55-0.75, with a 1.20x upscale for speed
Hi Res Fix For Forge, A1111: I truthfully have not messed around much, but i use the standard 4xUltraSharp, with a denoise of 0.55-0.75, and an upscale of 1.20x. no other settings are touched
NOW AS FOR THE LISENCE, ITS A SUPER STRICT ONE THAT STATES.... just kidding, do what you want, the amazing creators of the models i based this model off of dont push a disgusting agenda
although, if you make something cool with the model, i would love to see it. not nessacery, but credit would be nice, mainly just so i have a way of finding the cool shit yall make.. and dont be weird or evil, your responsible for that
give the creators listed above buzz or a comment, like, money, mouth kisses, idk, something that helps
P.S. this will be on all of these places under my BakaBlitz name and model name: Civit.ai, Salt.ai, Tensor.art, SeaArt, HuggingFace, and any other place you ask for thats not super weird lmao
Description
Massive thanks to @NicoGJ and @BetterThanNothing for their amazing models Toonify and Prancing Invasion Ribbon Wave.
ABOUT V2
I was a small bit dissapointed with The Idiotic Mix. i release models for myself first and foremost, and i release my models with the goal of super simple prompting, with great output. i found that prompting for V1 though was worse than my other models. prompt following wasnt as great as it usually is, and it was a bit harder to achive some of the things you can on my other models. plus aspects of the image, like eyes and fingers were inconsistent. so i made this version to fix all of that. you wont catch me making much more V2's. i will only put out stuff that i would use and do use
V2 FIXES!!!!!!!
-Anatomy is better. specifically eyes, as well as fingers. i personally had issues with getting good eyes on V1, and i would get mushy fingers often, but this is now as good as my other models :)
-Prompting is WAYYYYYYY better now. this was my biggest annoyance with V1, i found that it was harder than my other models to get the concepts that they are known for. if you looks at V1's showcase images, you will see that prompts are on average bigger than my other models. this is definetily so much better now. prompt following, ease of prompting, etc. its all really great now
-basically everything is just better overall. its fun to use. hope yall enjoy
Settings. All of the showcase images you see are completely stock with no post editing, no fancy comfyUI nodes, no controlnet, no inpainting. the only "post" thing i do is a latent upscale (hi res fix in forge and A1111). ive used ComfyUI (stock nodes), and Forge for the images seen, and copying the image into comfy/forge should give you the exact shit i did. i have also used the model in Fooocus, as well as sites like SeaArt, with no problems.
also, most images were first or second rolls, and i always gen 2 images at a time, generally all of the images shown were made in around 4-8 total images
for transparency sake i figured i would mention these. i can be confident in saying you wont need to reroll much due to awful anatomy, incoherent images, or artifacts
Steps: 30 or more, never tested under that, but im considering a lightning version of this. lemme know if u want that
Sampler: any of the DPM's with karras lol, im sure others will work fine
Negative Prompt: just used a copy paste i made years ago, as seen in the preview images, big negatives wont effect quality negatively, and small ones proly wont decrease quality much idk, just put watermark tags in there
Latent Upscale For ComfyUI: i use bislerp at the same Ksampler settings as your base image, and a denoise of 0.55-0.75, with a 1.20x upscale for speed
Hi Res Fix For Forge, A1111: I truthfully have not messed around much, but i use the standard 4xUltraSharp, with a denoise of 0.55-0.75, and an upscale of 1.20x. no other settings are touched
NOW AS FOR THE LISENCE, ITS A SUPER STRICT ONE THAT STATES.... just kidding, do what you want, the amazing creators of the models i based this model off of dont push a disgusting agenda
although, if you make something cool with the model, i would love to see it. not nessacery, but credit would be nice, mainly just so i have a way of finding the cool shit yall make.. and dont be weird or evil, your responsible for that
give the creators listed above buzz or a comment, like, money, mouth kisses, idk, something that helps
P.S. this will be on all of these places under my BakaBlitz name and model name: Civit.ai, Salt.ai, Tensor.art, SeaArt, HuggingFace, and any other place you ask for thats not super weird lmao