RainbowSweets

Goal

I’ve been mixing for a long time and this is the first one I think is worth sharing.  My goal was to create something bright and colorful but not so much that it washes out details, easy to prompt and get lots of details, clean crisp lines, and fun.  This model can produce a wide variety of content and while it can produce NSFW images, I’m not primarily focusing on that.  This model does well with Loras, I show a few examples but also just wanted to show what the base could do.

My Settings

Feel free to find your best settings, these are just mine.  I didn’t use ADetailer or anything fancy, but feel free to use them.

VAE: I created my own and baked it in.

Sampling Method: DPM++ 2M Karras, Restart

Steps: 25 – 30

Image Sizes: 512x512,512x640,512x768

CFG Scale: 6-8

Clip Skip: 1

HiRes

Upscaler: 1x-ITF-SkinDiffDetail-Lite-v1 or R-ESRGAN 4x+ Anime6B

Hires Steps: 15-20

Denoising Strength: 0.6

Prompting

Can handle both simple and more complex. Masterpiece and High Quality aren't typically needed from what I've seen but use whatever you like. Something as simple as 1girl can show great results with details and variety.

Negative:

Highly recommend Async’s Negatives, Async's Negative - v11 | Stable Diffusion Embedding | Civitai

These are just examples of what I’ve used, but I’ve tried many Negative prompts that have good results.

an11, an10, worst quality, low quality, multiple limbs, multiple legs, bad anatomy, NegfeetV2, bad-hands-5, badhandv4

an10, (sketch, watermark, signature, worst quality, low quality, large head, extra digits:1.4)

an11_light

Model Mixes

Thanks so much to all the models out there that keep mixing.  Here are some of the models I’ve mixed to get this model:

·         TMND-Mix - TMND-Mix_SP-RAINBOW | Stable Diffusion Checkpoint | Civitai

·         Sweet Sugar Syndrome - v1.5 | Stable Diffusion Checkpoint | Civitai

·         Yotta Mix - v6.0 | Stable Diffusion Checkpoint | Civitai

·         Kakigori - Kakigori_v3.0 | Stable Diffusion Checkpoint | Civitai