modelle/PicX_real - 10

PicX_real - 10

|
7/1/2024
|
8:44:47 PM
| Discussion

Empfohlene Negative Prompts

3d,cartoon,illustration,bad anatomy,deformed,ugly

Empfohlene Parameter

samplers

DPM++ 3M SDE Exponential, DPM++ 2M SDE Karras, DPM++ 3M SDE Karras

steps

24 - 30

cfg

4.1 - 6

clip skip

2

resolution

512x768, 600x904, 1024x768, 640x960, 768x1024

other models

picX_real (07919b495d)

Empfohlene Hires (Hochauflösungs-) Parameter

upscaler

4x_NMKD-Siax_200k, 4x-UniScale-Balanced [72000g], R-ESRGAN 4x+, ESRGAN_4x

upscale

1.6 - 2

steps

10 - 24

denoising strength

0.2 - 0.43

Tipps

Describe your desired main style, scene, details, effects, and artistic techniques for best results.

Ersteller-Sponsoren

support me on Ko-fi / Boosty

For business inquires, commercial licensing, custom models, and consultation contact me fp16guy@rundiffusion.com

The only authorized generation services:

RenderNet.ai Tensor.art

Model originally created by fp16guy


This model is less boring-rough and more diverse than the "ultra-realistic" models, and less animesh-asian and more realistic than the "soft-realistic" models. This makes it more balanced, flexible and gives the characters a softer appearance without losing the realistic part.

The flexibility of the model means that a basic understanding of how to write a prompt is required - describe desired main style, main scene, details, effects and artistic techniques, and in response it will show something pleasant. Everything else is traditionally on the preview.

Zurück
TI EasyNegativeV2 Textual Inversion Embedding - v2
Weiter
Add More Details Detail Enhancer Tweaker LoRA - v10

Modell-Details

Modelltyp

Checkpoint

Basismodell

SD 1.5

Modellversion

1.0

Modell-Hash

07919b495d

Ersteller

Diskussion

Bitte log in um einen Kommentar zu hinterlassen.