MENU

Introducing 4 Models and 9 Versions Optimal for Icons, Comics, and Anime – Including FlattQ | Stable Diffusion

This site uses affiliate links.
目次

Welcome to the World of RSTR, a Treasury of Heart-Melting Characters

She may seem a bit sassy, but deep down, she’s incredibly lonely.
Knowing that only cold replies come back, yet for some reason, you just can’t stop thinking about her.
Did you have such a girl in your surroundings?
RSTR’s models create one after another such adorable girls, letting you relive those stirring emotions once again.
Please take your time and enjoy their charm!

RSTR’s Newest Versions

The prompts were:

“1 girl” -> “wearing a casual cloth” -> “wearing a swimwear”

and I rotated through those. The only negative prompt was including the “EasyNegative V2” embedding.

I did not use any face refinement functions like ADetailer whatsoever.

Coquett

This model is characterized by its cute, sleepy eyes. It is unique, but I recommend trying it out. I recommend using the Euler sampler and Hires.fix. VAE is not included, so use your favorite one. The main models include “Petitcutie_v2.0” and “PuniMIX 4.0”.

(Summary of the site)

“This is an advertisement.”

FANZA Illustration Collection (R18)

FlattQ

Flat-tone JitQ

This model is a more flat and cartoon-like adjustment of JitQv3. It is designed for simple outputs and is not suitable for complex illustrations or beautiful landscapes. It includes VAE, but you are free to use it or not. The main models include “JitQ_v30” and “KawaiiTech-20”, and you can also fine-tune it.

(Summary of the site)

JitQ

Models used:

PetitCutie_v1.5
qteamixQ_omega
cuteyukimixAdorable_echodimension
Imperishable_alice

(Quoted from the site)

〈AD〉

PetitCutie

This model was created with the aim of cute and dynamic compositions. v1.5 is a version that applies LoRA to v1.0 and is adjusted based on the premise of using LoRA. Basically, we recommend using v1.5, but v1.0 is also suitable when flashy compositions are unnecessary.

(Summarizes the site)

RSTR’s Old Version

JitQ – v1.0

〈This is an advertisement〉

JitQ – v2.0 NOTJit

JitQ – v2.0

PetitCutie – v1.0

PetitCutie – v1.5

All Illustrations Are Commercially Usable

Some illustrations may have incomplete details such as hands and fingers. We kindly ask for your understanding.

When using these images, please feel free to crop or modify them as needed.

We encourage you to fully express your creativity and use these illustrations with a free and imaginative approach.

The illustrations featured on this site are available for commercial use.

Stable-Diffusion Settings for This Generation

The following are the settings used for this generation of Stable-Diffusion images:

  • Negative prompt: EasyNegativeV2
  • Steps: 40
  • Sampler: Euler a
  • CFG scale: 7
  • Seed: 3661322343
  • Size: 512×512
  • Model hash: e488852acc
  • Model: jitq_v20Notjit
  • VAE hash: 0b204ad0ca
  • VAE: klF8Anime2_klF8Anime2VAE.pt
  • Denoising strength: 0.32
  • Clip skip: 2
  • ENSD: 31337
  • Hires upscale: 1.5
  • Hires steps: 20
  • Hires upscaler: 4x-UltraSharp
  • Eta: 0.667
  • Version: v1.7.0

Additional note:

The upscaler “4x-UltraSharp” seems to be popular, so I’m using it without really understanding why. (laughs)

目次