Description
FAQ
Comments (7)
The images all come out kind of squashed, like a stretched 4:3 image. Is there a workaround for this?
it was trained on 1920x1080 images maybe try width images
Unfortunately, generating wider images doesn't solve the problem at all.
The images used for training have the wrong aspect ratio, so at present the only way to fix the aspect ratio seems to be with an image processing software like Photoshop.
We'll have to wait for someone to create a LoRA that can change the aspect ratio of images:-)
Maybe try AutismSDXL it was trained on this checkpoint and try it locally it looks nice locally// This website generate different images for model trained in kohya locally// and generate different images for model trained in this website
@OssanGTR go see this images i attach them into model
Can you do Guts next?
This lora have guts i think :)
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.