DarkPhoenix 3D版本v1.2 (ID: 226913)

First model merger test I did, credits to the models below.

AbsoluteReality: https://civitai.com/models/81458?modelVersionId=132760

BeautyFool: https://civitai.com/models/101888?modelVersionId=134148

realspice: https://civitai.com/models/158734?modelVersionId=208629

For this model I recommend using some basic prompts like positive and negative.

Positive prompts: masterpiece, high quality, best quality, Detailed

Negative prompts: EasyNegativeV2, low quality, normal quality, worst quality, (bad anatomy)

Other combinations of negative prompts also work, as does using Fastnegative or BadDream (but I strongly recommend using EasynegativeV2, because of all my tests, this is the negative that works best without damaging the image too much in StableDiffusion)

Don't use Embedding NegfeetV2, I don't know why, but it ruins the image if you use this Embedding in this model

Recommended settings

Sampling steps: 15-30 (Only use more steps as a last resort for very complex images, as an excessive number of steps tends to ruin the image)

CFG Scale: 5-12

Hires. Fix

Upscaler: 4xNMKDSuperscale_4xNMKDSuperscale

Hires Steps: 15-20

Denoising Strenght: 0,20-0,40 (The more force, the cleaner the image, but the more the original quality is lost.)

Upscale By: 1,5 - 2,0

Adetailer:

Adetailer Model: face_yolov8n.pt

Inpaint denoising strength: 0,25 - 0,40

ADetailer steps: 15 - 20

Did you like the content? I would be grateful to receive a coffee

https://ko-fi.com/berserkfg

This model is open source for pornography, therefore pornographic images are at the image creators' own risk, and the model will not be held responsible for any image with heinous content.

描述:

I made some changes and I think it improved the eyes and colors, and it seems to have more details.

This time I generated all the images without using Lora, my idea was to show the model's capabilities without interference from any Lora.

训练词语: