Heathens

Another attempt at a full realism lora, however due to leaving PARTIAL text encoder on (like bare minimum) it DOES work on all models.

FOR SOME REASON THE SDXL VERSION GOT CORRUPTED -- we'll bring it back when we get a chance to offsite test it!

This is also meant to be a demons * Halloween type style. It's not perfect, there was some issues with tagging since I tried to develop my own local tagger and I failed miserable due to cloud compute. (KNX Tagger is archived unless someone wants to fork it and allow for local LLM users to use it, Gemini and Cloud LLM's are too expensive and rate limits are harsh on free tier)

✨ Recommended Settings

To get the best out of this mess, try these settings:

• Samplers: Euler or Euler Ancestral

• Scheduler: Karras

• Negative Prompt: Up to you, we use a mix of our embeddings, sometimes don't even use negatives!

• CFG & Steps:

  • A mix of anywhere from 5 to 7 for CFG

  • Steps is your choice, we do lower because we're lazy but some peopel go all the way up to 50! (Or more!)

• HiRes Fix:

  • Steps: 15

  • Denoise: 0.3

  • Upscale: 2x

• Upscalers: Rex, 4x_Remacri, or your personal favorite.

• ADetailer: Use sparingly. Due to potential VAE differences, it might occasionally produce grey squares.

• Embeddings: We've used some of our own custom embeddings to showcase character types in our examples, but no stylistic embeddings were necessary to achieve the core look.

• LoRA Compatibility: Chances are good that most LoRAs will work with our new models. However, be aware that while they will function, the result might not perfectly align with your expectations if the LoRA heavily influences style. It's a "What You See Is What You Get" situation with the base model's strong aesthetic – it'll likely produce a good image, just perhaps not the one you had in mind with a very specific LoRA.

🎨 Model Style & Inspiration

This Mix aims to capture the charm and the inspiration of the dynamism of YOUR imagination, and the expressive qualities of illustration. Whether you're crafting whimsical characters or vibrant scenes, this model is ready to explore with you!

🛠️ Merge & Training Details

This is a TRAINED lora, we rarely do fused/merged loras (but they're around!) - the backbone of this is a mix of unsplash and open sourced stock images plus synthetic data. We're trying to build up some loras that CAN be used on multiple styles, but are trained on mostly realistic data.

🙏 Acknowledgements

A huge thank you to the incredible community members whose work and inspiration contribute so much. Special mentions to:
Arthemy, Wai, Holostrawberry, FallenIncursio, RichyRich, JustTNP, Hoseki, CitronLegacy, Samael1976, Kairen92, ShadowxArt, Flyx3, YeiYeiart, LazyVisionary, and so many more! We try to train as much as we can ourselves, but the community's creativity is a constant source of inspiration.

🌟 About Ktiseos Nyx & Connect With Us!

Ktiseos Nyx is available for commissions! Find us and more of our creations here:

• Our Discord Invite Link: https://discord.gg/HhBSvM9gBY

• If somehow the original link DOESNT work use this one: https://discord.gg/MASBKnNFWh

• Earth & Dusk Media (Not AI) https://discord.gg/5t2kYxt7An

• Our website (WIP): https://knx.neocities.org/

• How to support: Patreon: https://www.patreon.com/earthndusk

• Membership: https://ko-fi.com/earthnicity

• Membership / Ko-Fi: https://ko-fi.com/duskfallcrew/

• Membership / Ko-Fi: https://ko-fi.com/OTNAngel/

• Buy Me A Pizza/Coffee: https://www.buymeacoffee.com/duskfallxcrew

• Merchandise: https://duskfallcrew-shop.fourthwall.com/

• PixAi Referral: https://pixai.art/referral?refCode=BRU9AUNA

• Rent Serverless GPU with VastAI: https://cloud.vast.ai/?ref_id=70354

• Railway Deployment Referral: https://railway.com/?referralCode=EQxw4P

(Text in model cards Copyright 2025, 0FTH3N1GHT Productions, Earth & Dusk Media & Ktiseos Nyx)