Tensor.Art
Create

#Halloween Epic Makeup Portraits

LORA
Original


Updated:

404

Shameless Plug: SDXL, Cascade, Upscaling, All My Models in a customized ai pipelines. Check out 🤫 https://www.diffusionaiart.com/

They are also exclusive for Tensor.Art only.

Thank you ❤️⭐

Hi! Back with another awesome LoRA. This model is configured to generate what couldn't have been achiavable before. With the curse of Halloween upon us, I wanted to generate something new, and exciting.

I call all the tensor art people to try and discover this lora. I put around 70 images in this data set and trained it over 10 hours.

Weight: 0.6-0.9

Aspect Ratio: 1024x or 768x

DPM ++ 2M SDE Karras and DPM++ 2M Karras gets the best results

Add lora to the Adetailer extension also!

Good checkpoints to use this lora with: CrystalClearXL, SDXL1.0, please share you generations if you test more.

Prompt suggestions: "Portrait of "subject" in makeup", "portrait of "super mario" man in makeup", "portrait of a woman in spiderman makeup," etc... Try using trigger words to further enhance the effect. However, try to keep it simple as this model is littlebit overtrained and further update will be coming soon.

*This lora is trained at high steps so its sometimes not so easy to prompt. Might Need detailed prompting.

*Results might look a bit dirty, fracted... This is result because training images were mostly scattered images like the outputs. To fix this you can use img2img sd upscaler extension with around 0.2- 0.25 noise ratio, it should clear the image up. I am already thinking of possible fixed, will be updating soon!

*Another problem is since it is trained at so high step count, it is hard to create anything other than portraits. It's not impossible, but rather harder compared to my other loras. Keep that in mind.

My other loras:

Magic Blender XL ✨(RealismAI)

https://**********/models/629272469309621506">https://**********/models/629272469309621506

DallEngine XL

https://**********/models/644131875037737631">https://**********/models/644131875037737631

Armored Core XL

https://**********/models/633413032531340165">https://**********/models/633413032531340165

Geometry in Everything (RealismAI) https://**********/models/630489388163468811">https://**********/models/630489388163468811

TransparentSkinXL Detailer

https://**********/models/627423885320645243">https://**********/models/627423885320645243

LoFi Art Style XL

https://**********/models/626017493984561756">https://**********/models/626017493984561756

REALismAI XL

https://**********/models/625723890020160772">https://**********/models/625723890020160772

Coloring Book XL Dominator

https://**********/models/625574523942624468">https://**********/models/625574523942624468

Ogkai <3 pls pls support my channels <3

www.instagram.com/ogkai_11

Version Detail

SDXL 1.0
12800
10
don't forget to use trigger word (aw0k halloween makeup) at the beginning of the prompt, if not output becomes less stable!! Best checkpoint i tested is: Crystal Clear XL - 1 Weight: 0.8 tested DPM++ 2M SDE Karras (min 30 steps, more steps give more detail) Don't forget to use Adetailer too (+add lora to the adetailer also!!) Example prompt: (aw0k), "portrait of a man in spiderman makeup", portrait of a woman in makeup, moonlady paint", Negative: text, watermark, low-quality, signature, moiré pattern, downsampling, aliasing, distorted, blurry, glossy, blur, jpeg artifacts, compression artifacts, poorly drawn, low-resolution, bad, distortion, twisted, excessive, exaggerated pose, exaggerated limbs, grainy, symmetrical, duplicate, error, pattern, beginner, pixelated, fake, hyper, glitch, overexposed, high-contrast, bad-contrast <lora:negativeXL:1> Seed: 2034254805 As I tested crystalclearXL and base 1.0 model gives nice results.

Project Permissions

Reprinting is strictly prohibited

    Use Permissions

  • Use without crediting me

  • Share merges of this model

  • Use different permissions on merges

    Commercial Use

  • Sell generated images

  • Use on generation services

  • Sell this model or merges

Comments

Related Posts

Describe the image you want to generate, then press Enter to send.