dreamScan

Version 1.2

1. Improved aspect ratio support.
2. It should be a little better at understanding locations and seasons.
3. It should be better at generating bodies.

Try to keep the longest side to max 512px on initial render.

IMPORTANT!

With v1.2 there is a change in prompting. Instead or solo or 1girl, try some of these:

dreamScan, scandinavian woman, natural beauty, nordic, blonde hair, outdoor, forest, casual, smiling, natural colors, candid, soft lighting, high detail, muted colors, suburban, city, parking lot, gas station, hospital, store, home, apartment interior

Read the terms at the bottom of this post, please be responsible, or the model will be taken down and I will not release a future version. Thank you all for the support so far.

Version 1.0

A Lora used to create realistic Scandinavian faces and backgrounds.
Trained using and for Realistic Vision 5.1 (VAE) but works on the Hyper version.
See my generated images for prompt suggestions.

Suggested settings

Lora strength: 0.6 - 0.8 (Experiment!)

Clip skip: 1

Sampling: DDIM

Schedule: SGM Uniform

Sampling steps: 22 - 28

CFG: 2.5 - 4.5

Additional (optional)

ADetailer: mediapipe_face_mesh_eyes_only

Enjoy and be responsible!

Terms and conditions

I strictly prohibit the illegal use of this model, including the creation of inappropriate or indecent content.

Please do not repost or re-upload this model on other websites—I do not grant permission for this. If you come across unauthorized uploads of my model, kindly report them to me.

If I find that this model has been used unlawfully or distributed without my consent, I reserve the right to take it down and will discontinue any further development of similar models.

You are solely responsible for addressing any intellectual property concerns related to the content generated with this model. It is your duty to ensure that its usage complies with all applicable laws, regulations, and the terms of this service. Unless explicitly allowed by law, this model may not be incorporated into third-party software or services, nor may any connection between the generated content and this service be disclosed to others, whether directly or indirectly, explicitly or implicitly.

This model is strictly prohibited from being used for commercial purposes, including but not limited to text-to-image, image-to-image processing, or face-swapping. Any violation of these terms will result in legal accountability, and the author reserves the right to seek compensation for any resulting damages.