Home Artists Posts Import Register

Downloads

Content

Patreon exclusive posts index

Join discord and tell me your discord username to get a special rank : SECourses Discord

I started working on preparing an amazing prompt list for Stable Diffusion self trained models.

Why this is important? Because many of the prompts have strong face influence which will cause bleeding of face. Therefore, specially curated prompts are necessary to get best results with person trained models.

The v1 prompt list.pdf file currently has 58 unique prompts but I have generated images for 4500 more styles and I will pick and add best of them to the PDF file hopefully.

I am also sharing generated images as v1 prompt images.zip attachment so you can directly drag and drop them into the PNG info and very quickly use them. Don't forget to extract zip file.

Please read the attached PDF file carefully.

Also some prompts are using styles.csv so if you put it into the Automatic1111 folder they will be automatically selected.

Finally all prompts are attached as a v1 all prompts.txt so that you can use them as a batch in Prompts from file or textbox of Automatic1111 interface.

New prompts not added yet

  • render of ohwx man character wearing a suit , 3d, cgi, blender, computer graphic , Volumetric Lighting, ray tracing, shadow mapping
  • Realistic negative prompt : cartoon, drawing, ugly, deformed, noisy, blurry, low contrast, realistic, 3d, cgi, render, anime, blender, graphic, drawing, digital art, sketch, line art, disfigured, mutated, abstract, 2d, minimalist, vintage, distorted, glitch, manga, Blurred, Hazy

Here some example pictures

Comments

Ec Jep

Thx for sharing what you’ve learned. Saves me so much time! Is it worth using two rtx cards (4090 24 gb and 3060 12 gb)? Trying to figure out if I should upgrade power supply to handle both

Manpreet Singh

Thanks for sharing this prompt list. Amazing work! Any plan to include prompts for professional headshot photos? With various backgrounds, clothing etc.? That would be super useful as well.

Felix Rockwell

Thank you, I used the v1 prompt images.zip at my trained lora from Kohya, but it is changing my hair style, do you know why? On your images it seems you have the same hair as in your real life, but for me, it's giving me these modern hair style which I don't want.

Furkan Gözükara

i would suggest to train dreambooth with our latest config : https://www.patreon.com/posts/very-best-for-of-89213064