Verified: 3 months ago
SafeTensor
Type | |
Stats | 62 73 |
Reviews | (10) |
Published | Oct 17, 2024 |
Base Model | |
Training | Epochs: 5 |
Usage Tips | Clip Skip: 2 Strength: 1 |
Trigger Words | SagePXL |
Hash | AutoV2 9CFB1B44A0 |
This is Sage, one of the characters from the adult visual novel "Being a DIK" (you must be of legal age to play this game).
Lora seems to work better with the prompt "3D render" but feel free to try it without.
Lora seems to work well with a strength setting of 1 or 0.8.
The example image uses the following method: I created five images with 30/40 sampling steps, size 512x680 with a CFG scale of 7. Then, I selected one, sent it to image to image, and generated it with 40 sampling steps, sized 1024x1360, with denoising strength between 0.2 and 0.5. Additionally, I turned on ADetailer for eyes only.
I plan to enhance the eyes and background flexibility in the next version.
I hope you will enjoy using Lora. Don't forget to rate it, leave comments, and share your creations! Your feedback is always appreciated! Thanks!