This is a oficial release of my HC mix.
Read the model description to recreate the images, any doubt, join my discord channel or send to my instagram account.
You can found more images at the model's discord.
You can follow my on instagram
This is a HARDCORE version of kaelvinSmix. Include some @weights of uber, abyss, basil and chillout .
My custom merge with some chinese model's, and some nude model's . The idea is not to generate something completaly new, but some model that can be use to create a big variety of images.
I don't own any model used in this merged, so:
All rights goes to them.
I just upload it here, for those who like to test it.
This model has restrictions
You agree not to use the Model or Derivatives of the Model:
In any way that violates any applicable national, federal, state, local or international law or regulation;
For the purpose of exploiting, harming or attempting to exploit or harm minors in any way;
To generate or disseminate verifiably false information and/or content with the purpose of harming others;
To generate or disseminate personal identifiable information that can be used to harm an individual;
To defame, disparage or otherwise harass others;
For fully automated decision making that adversely impacts an individual’s legal rights or otherwise creates or modifies a binding, enforceable obligation;
For any use intended to or which has the effect of discriminating against or harming individuals or groups based on online or offline social behavior or known or predicted personal or personality characteristics;
To exploit any of the vulnerabilities of a specific group of persons based on their age, social, physical or mental characteristics, in order to materially distort the behavior of a person pertaining to that group in a manner that causes or is likely to cause that person or another person physical or psychological harm;
For any use intended to or which has the effect of discriminating against individuals or groups based on legally protected characteristics or categories;
To provide medical advice and medical results interpretation;
To generate or disseminate information for the purpose to be used for administration of justice, law enforcement, immigration or asylum processes, such as predicting an individual will commit fraud/crime commitment (e.g. by text profiling, drawing causal relationships between assertions made in documents, indiscriminate and arbitrarily-targeted use).
Dreamlike Diffusion
License
This model is licesed under a modified CreativeML OpenRAIL-M license.
You can't host or use the model or its derivatives on websites/apps/etc., from which you earn, will earn, or plan to earn revenue. If you want to, please email us at [email protected]
You are free to host the model or its derivatives on completely non-commercial websites/apps/etc. Please state the full model name (Dreamlike Diffusion 1.0) and include a link to the model card (https://huggingface.co/dreamlike-art/dreamlike-diffusion-1.0)
You are free to use the model or its derivatives for commercial purposes in teams of 10 or less
You can't use the model to deliberately produce nor share illegal or harmful outputs or content
The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license.
Description
Release the first official version of KSMHC.
Add ChilloutMix to the mix.
Remove 0.15 of uber.
decrease Abyss by 0.5 and add Basil by 0.15.
FAQ
Comments (25)
What prompt are you using for the cum covered ones?
Bukkaki, Bukkake, between 1.21~1.41, Or try add the bukkakAI Textual Invertion, You can download it here at the CIVITAI
@kaaelvin992 What setting is the 1.21 - 1.41 one?
On the prompt, use the tag (bukkake:1 21)
that looks interesting! but is there a chance to have it as safetensor?
You can download the alpha 0.2. It's the same version of 0.2.1 but a safetensor file.
Can it produce full hardcore (penetration) well?
just the HC+, it lll be release first at my discord for testing tomorrow, than here after a week
Hi, can you create new invite to your discord? (old not work)
https://discord.gg/76EeC7braC, here it s! ty for the tip.
This is weird... I tried recreating the images you posted here and copied the exact same settings, but the generated pictures are different and also everything has very muted colors.
With vae are you using?, Also there's some versions of the KSMHC. Are you using the same version of the image? If you still can't reproduce, call me on discord, i ll help you there
Hi, thank you for responding :)
I'm not using VAE, and running version alpha 0.21 (The most recent one)
Maybe your'e using a lot of embeddings I'm missing? Are you using Automatic1111's repo?
How can I find info about using VAE in stableDif?
Hi. I can't figure out which weights to use.
Definitely on the same boat as others here, I tried recreating a few to see what would come up but everything seems rather washed out and generally not matching. Same thing happened with the SMix model.
Is there a VAE you recommend for it? Or am I missing something?
Which one is you want to recreate?
Sorry that I didn't mention I used both txt2img and img2img with same prompt and different seeds.
details:
https://www.reddit.com/r/WaifuDiffusion/comments/10zgopv/cyberpunk_wallpaper_kind_of/
I thought the waifu diffusion vae is good for high saturation and colorful pic.
@cyl Basically I tried recreating several to see how flexible I could make it but overall the main issue I'm getting is essentially washed out colours with low definition (as in poor edges, very twitchy)
From the prompts I tried the two shibari plus the doggystyle ones you had there.
Witch UI are you using? @RykeWollf13, and about the vae, use the vae-ft-mse-840000-ema-pruned. If you are using the auto1111, there's a know bug about the auto select vae with the same name (there's a checkbox at the config tab, try to uncheck this, if this is the case)
@kaelvinsmix Yep that did the trick, for most models I have the VAEs with the same name as the models, not realising I could actually select them. The moment I chose the suggested one, all perfect, cheers!
Great examples, but my experience is disappointing.
I couldn't get anything close or even similar to the examples, although I used the same prompt string, the same number of steps, and the same seed. The result turned out to be completely different both in colors, light / shadow, and in plot (with many body artifacts). I think that in order for us to achieve the same result, the author of the model needs to include everything necessary like VAE in the model and give the necessary detailed description of how this can be achieved, instead of a description of the license. I'm using DiffusionBee and as far as I know I can't use external VAEs, although I could be wrong because I'm still new to this.
All images are created in 512x768 than hi res to 1024x1536;
Work better with skip clip 2.
Vae changes between 8400, and f8/ anyfruit
For hi res on low gvram, create the image, move it to img2img, copy the seed, change de size to 2x the original value, and work with denoise around 0.25 ~ 0.5. It's not a precise math since for each sample i work on some way, but if you can make the core image in low res, you can uspcale in img2img or using original hi-res check box
How about pruned version? plz
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.





