I made a video on how to achieve the same quality or just repeat my art
P.S. If you want to support me, then you can subscribe to my Patreon and get builds earlier than they come out here (versions on patreon will always be newer by one than here) This will help me a lot in updating the video card over time to make models even better and faster (the main thing is better)
The first lora where I did the preprocessing right away, and not after several versions. With the help of neural networks, I improved all the art, removed the background noise, removed watermarks and other elements of the text, except for the background, made a complete selection of prompts manually! In the end the result was worth it) The purest art.
Thanks to the high quality of the images, you can achieve detailed backgrounds in the same style, which was not in my past works
LORA weight from 0.2 to 1 depending on your goals, 1 - closer to the original style, there are no restrictions here. I use 0.6-0.7
CFG Scale - 3.5-5 (According to my personal observations, 4 is the best choice)
Use model Abyssorangemix3aom3_aom3a3.safetensors
just use hires fix as much as your video card is capable of, so as not to give an error, with 8gb of video memory it is 512x768 with hires fix 2, upscaler Latent (nearest-exact) or just Latent (if there are problems with the image), the picture will be much more detailed.
Trigger Words: Faye Valentine
Combined with my other LORA
How to use LoRA's in auto1111:
Update webui (use git pull
like here or redownload it)
Copy the file to stable-diffusion-webui/models/lora
Select your LoRA like in this video
Make sure to change the weight (by default it's :1
which is usually too high)
*Information taken from Lykon
Please post your work with or without comments, it will help me improve. Thanks!
If you like my work, click on the heart above, I will be pleased :3
Type | LORA |
Downloads | 3,120 |
Uploaded | Aug 30, 2023 |
Base Model | SD 1.5 |
Trigger Words | faye valentine |
Rating | 5.00 (9) |