This is a cosine merge of EnvyMix 1.2 and Pooda-Beep Mix Illustration-Comic that brings a bit of Pooda-Beep's boldness (and ability to draw good hands) to an anime fusion style. It has a unique personality that's not quite like either parent. It's also quite stable with LoRAs and Lycos.
Description
FAQ
Comments (17)
Aww yisss 😍😍😍 -Luan
Very interesting, do you also mix your various style lora and theme lora (like witch / villain etc) into the mix? Or are those still complementary rather than built in directly
Those are still complementary. I'll have to do a test and see what happens when I mix them all in together, but I have a feeling that all of them will affect everything and you'll end up with a mishmash. A cool looking mishmash, though.
Yes sir !
Envy style is very trendy atm,so i am picking em n huntin em like a hot pancakes from a frying pan i swear.
Where can I find the DPM++ 2M Karras Sharp v1 that you use in your samples?
In this thread here:
https://github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/8457
Go to the top comment and click the edit button. I'm using the modification from the most recent version of the comment from March 9th.
Later on, they made some changes to it that IMO were for the worse.
@_Envy_ did you rename it to sharp v1? when I make those changes it is called "DPM++ 2M Karras Test"
@x_mkiv I did, yes.
I loved your model! But i can't seem to get to the same results in these anime like pictures, without all the high contrasts and flashy things. I'm stuck with the semi-realistic portraits which was what i was trying to runaway from
Hey, sorry I didn't see that earlier. Can you post one of your images along with prompts, what VAE and loras and such you're using, your hires fix settings, your adetailer settings, etc (and let me know if any of those things don't mean anything to you, because I can help you with that stuff as well).
Sorry, i didn't see your reply either. I downloaded the vae you recommended, but i'm kinda new and i have no idea what an adetailer is. I'm gonna take a look and also post a pic here
I feel like I'm missing something obvious, but I cant for the life of me get either the fp16 or fp32 models to work. They output purpleish colored images broken up into squares. Its not a refiner or a VAE issue, and I've tried redownloading in case it was a corrupt file.
What else should I check? I am using ComfyUI.
Did you manage to solve it? The same thing happens to me in stable diffusion forge.
@YamKisHori If I recall correctly my issue was due to using the wrong dimensions. I think I was using SD1.5 dimensions with an SDXL model.
I can't get it to work in any way, all the images show me a square of some flat color or just noise.
in a very beautiful model and I can't use it D,:
I'd re-download it.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.