Concept LoRA for meme from the movie "Predator".

For NAI Vpred / Epsilon1.1 / IL (WAI).

TAGS:

epic handshake \(meme\), looking at another

optional:

emphasis lines

helping tags:

upper body, cowboy shot, hand up

Use forge couple for forge / reforge if you use character lora. Usually if characters are known by checkpoint forge couple is not so necessary.

Use 1024x1024, 1152x896, 1216x832 resolutions for lora (it is more trained for wide pictures or 1:1). Don't use portrait resolutions.

EXAMPLE OF FORGE COUPLE PROMPT

How I made this picture: https://civarchive.com/images/96851640

You need to write prompt in next order:

1. aesthetic tags, general composition, how many characters (2girls, 2boys, etc), what they are doing. Also you need to have activation tag and lora at this line.

2. Line 2 for character 1. Activation tag (Usually name), apperance, outfit.

3. Line 3 for character 2. Activation tag (Usually name), apperance, outfit.

You need to divide these prompts by Enter. No other symbols or commands like BREAK.

Then enable forge couple and spread zones for 3 lines: all picture for line 1, left part of screen for line 2, right part of screen for line 3. They can cross each other if it's necessary in zones where characters touch each other.

If you have only 2 lines in forge couple press "add" (blue button).

Screenshots (zones for each line):

NoobAI-XL quality prompt:

Put "masterpiece, best quality, good quality" in positive and put "low quality, worst quality" in negative.

Use euler, cfg 4-6. Don't use higher CFG!

Guide about NoobAI-XL Epsilon.

For NoobAI-XL V-pred:

Check my guide about V-pred image generating!

The best model to use V-pred is Cyberfix! But pure NoobAI-XL V-pred is also good.

For Illustrious mixes and NoobAI-XL Epsilon: use NoobAI-Xl Epsilon (especially mixes based on NoobAI-XL Epsilon!).

Upscaler / adetailer settings:

Hires.fix: Upscaler foolhardy_remacri (download and put in "ESRGAN" folder in "models" if you are using automatic1111 or forge).

Hires steps: 12-16

Denoising strength: 0.2-0.3

Adetailer: denoising strength 0.25-0.3. Detection model confident threshhold: 0.7-0.8.

Best sampling methods (with 25-40 sampling steps for txt2img):

• euler a SGM Uniform, CFG 5-6;

• euler ancestral CFG++ SGM Uniform / Simple, CFG 1.2-1.5.

You don't need any DPM++.

I will be glad to receive feedback from you and a demonstration of your art with my LoRA!

https://tensor.art/u/604375505701151021