To begin with, English is not my mother tongue, so I apologize in advance if there is any mistake.
I think this checkpoint can be unstable sometimes.
The images (character faces) I posted with this model have been inpainted at least one time.
The images are use multiples embeddings in the negative prompts so may be different if you don't have these embeddings.
Sampling method: DPM++ 2M SDE Karras
Sampling steps: 30
Upscaler: ScuNET PSNR
Hires steps: 20
Denoising strengh: 0.51
Please, don't use restore face.
I don't have done a lot of testing, and i won't (well Diablo IV is coming XD).
Please, please, experiment and show me what my checkpoint is capable of !
Description
Pruned version
FAQ
Comments (26)
Please don't worry about your English, it's good. And you know more languages than I do! :)
Haha, thank you !
A question, why in the process the image is a colorful one and after the image is finished it comes out gray, pale and without color?
I'd say something in your prompt ? Try with "greyscale" or "monochrome" in your negative prompt
also, check out using VAE. Don't know which ones work with this checkpoint. Check out using "VAE's" Google it you know" using VAE in Stable Diffusion" and you should be on the right path. in fact check this out here on civitAI
https://civitai.com/articles/72/fix-your-colors-vae
explains much, be well
I really liked this model! Can I use images created on this model for commercial use?
what it's the difference between v1.0 and v1.0 pruned ? and waiting for v2 <3
Pruned version is lighter. If you don't intend to do training with the model, Pruned version is the one you want.
Is there any chance for an SDXL 1.0 version?
By far my favourite model! Any plans to update?
Thank you ! I don't think so, but never say never, idk
I've said it before, but this is the best model on the site imho. so freakin aesthetically pleasing renders.
Thank you =)
Can you make this available for the onsite generator?
Do you have any plans on making a sdxl/pony version of this great model?
Not really, but maybe some new models. Maybe.
@ForgottenLord Or maybe SD 3.5 now if the finetuning is going to work better than with FLUX's rigidness.
Anyway, still one of my favorite models.
@barefooter1234 Thank you ! =)
Your 'Animation - 2D' mix inspired me to begin generating almost two years ago. Now, I use this model’s unique and rich style while inpainting and upscaling to make many of my favorite new images. Thank you!
Thank you very much, it makes me happy
Still one of the best checkpoints on the site. Really hoping for an SDXL/pony or other variant of an updated model. Hope you get the chance!
Thank you very much !
I may post new models (for exemple the one i used to create my profile picture), but it won't be updated versions or variants of my existing models.
@ForgottenLord If possible, can we have the data set, or some lora for Sdxl/IL for this style? I have never managed to replicate it with any other model and I can say this is hands down the best style. Would be willing to pay/help for it.
@ZeroEight This model was created by randomly merging models/loras. It was created years ago and i have honestly no clue about the models it's made of.
I don't know how to make a lora but is it not possible to generate images with this model then to train a SDXL/ILL model using these images ?
@ForgottenLord Yup, I might just create a lora at one point. Will keep you updated if I start or even succeed.
@ZeroEight Thank you and good luck !
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.