models/Custom Mix - Ash Milkshake [Illustrious | Noob] - Milkshake V4 FINAL ver.

Custom Mix - Ash Milkshake [Illustrious | Noob] - Milkshake V4 FINAL ver.

10/22/2025

11:11:40 PM

Related Keywords & Tags

civitai generator,custom mix - ash milkshake [illustrious | noob],illustrious,image generation,kumquatmcgee,lora,milkshake v4 final ver.,noob lora,noobai vpred,ntr mix,stable diffusion model,style,vpred parameters

A girl with long white hair wears a spiked barbwire wreath and a white blindfold dripping black blood, showing sharp fangs and a sadistic smile, dressed in a sundress with freckles and a spiked collar.

Recommended Prompts

Furry

<lora:MilkshakeMix_Test_v10_Merged5:0.85>, 1girl, best quality, newest, BREAK masterpiece, very awa, highres, absurdres

Recommended Negative Prompts

worst quality, low quality, bad quality, lowres, oldest, mid, watermark, artist name, patreon logo, patreon username, signature, worst aesthetic, monochrome

Recommended Parameters

samplers

Euler a

steps

cfg

3 - 4

resolution

832x1216, 1152x2016, 1248x1824, 1216x832, 960x2304

other models

MilkshakeMix_Test_v10_Merged5 (f94ab7211145), noobaiXLNAIXL_vPred10Version (ea349eeae8)

Recommended Hires (High-Resolution) Parameters

upscaler

4x_foolhardy_Remacri

upscale

1.5

steps

denoising strength

0 - 0.4

Tips

Turn on advanced settings when selecting the model on Noob or Noob-derivative models with the Civit online generator.

Use lora strength between 0.7 and 1.0, commonly 0.8-0.9.

Start prompts with "Furry" for better Anthro image results.

Version Highlights

The metadata in the initial images I posted all use a different Lora name. Do not copy that name it won't work. "MilkshakeMix_Illustrious_v4" should work fine.
Weight: 0.75-0.9. Feel free to go higher or lower if you like.
This is the final version. It took me too long. It took me months. I need help. It's not perfect, but it's as perfect as I'm willing to take it.

UPDATE: If you wanna run this model on Noob, or Noob-derivative models with Civit online generator, make sure you turn on advanced settings when selecting a model.

The Noob version seems to kinda work on the Civitai generator if you drop the CFG scale to ~3.0 (I'm using Euler a). It's not gonna have the same quality as genning locally since Civit doesn't support Vpred yet. Using the Civitai generator will result in oversaturated images.

A custom style based off of a bunch of artists/mix from NTR Mix, which I generated a bunch of images off of and trained for NoobAI Vpred.

This lora is trained using Vpred paramaters so please ensure you can run Vpred models. It works best on NoobAI Vpred v1.0, although it might not work on the Civitai generator because they don't support Vpred (yet).

Recommended lora strength 0.7-1.0 (I usually use 0.8-0.9). Don't forget to use "Furry" at the beginning of your prompt for better Anthros.