modelle/SDXL FaeTastic Details - v24

SDXL FaeTastic Details - v24

|
7/16/2025
|
11:49:15 AM
| Discussion

Empfohlene Negative Prompts

bad quality, low quality, blurry

Empfohlene Parameter

samplers

DPM++ 2M SDE Karras, Euler a, DPM++ SDE Karras, DPM++ 2M Karras

steps

40 - 45

cfg

5.5 - 7.5

resolution

832x1216

vae

SDXL-VAE_0.9.safetensors (63aeecb90f)

other models

SDXLFaetastic_v24 (07b985d12f), FaeTasticSDXL24 (603b6d615a)

Empfohlene Hires (Hochauflösungs-) Parameter

denoising strength

0.23 - 0.44

Tipps

Recommended usage strength for this LoRA is between 0.6 and 0.8 to avoid an overly strong effect.

Using 'Faetastic' as a trigger word is optional.

This LoRA focuses on enhancing details and saturation with a magical feel and is not intended for realism.

Open-source use and merges are encouraged with credit; commercial use via third-party generation sites without creator benefit is strongly opposed.

Versions-Highlights

testest

This LoRA is used to enhance images, giving them more details, increasing saturation, and making things overall prettier and magical feeling. Unlike the other FaeTastic Details I didn't bake it into my model, so you're able to freely use it with my newest version of FaeTastic.

I trained this LoRA on a bunch of my 1.5 generated images that I felt were high quality with resolution, color, and details... a lot of them involved my TIs which I think put a lot of 'magical' effect into it. This LoRA IS NOT realism focused.

I recommend a strength of .6-.8 for this model, otherwise, things can get a bit too strong....unless you want that effect. You can use Faetastic as a trigger word, but it doesn't really need it. :]

As always if anyone has any questions or needs troubleshooting with Civitai or my resources, you can always comment here or message me on Discord, I am in the Civitai discord.

Because I believe in the open-source spirit of SD, anyone is free to use this model or any of my LoRAs to merge or use as they see fit in an open-source sense, including selling any images or using them in personal projects. I would be extremely grateful to be credited for any merges or uses of my resources.

However, I am very opposed to having my model and resources I personally train and test and spend hundreds of hours on put up on third-party generation sites for their monetary gain where I receive nothing in return. ❤️

Zurück
experimental - m100-style - moody-fog01
Weiter
Motimalu - All Rated (Illustrious) - v1.0 [Illustrious]

Modell-Details

Modelltyp

LORA

Basismodell

SDXL 1.0

Modellversion

v24

Modell-Hash

1cf798aca8

Ersteller

Diskussion

Bitte log in um einen Kommentar zu hinterlassen.