A simple model for generating LEGO.
Notice to add the trigger word "lego".
Recommend using hires fix.
Description
FAQ
Comments (9)
No prompt examples?
I'm sorry, I don't know why the image I uploaded cannot read the prompt , but you can download the image and drag it directly into the webui, which can read the prompt.
very convenient to generate toys with charactor lora
Your images are probably not PNG format, if you make a few in PNG format then Civitai can read the meta data. It has issues with webp meta data.
Thank you for your answer. My images are PNG format, but they still can not be read.
@rainn00 No they are not... Try to save them or rightclick and open image in new tab, you'll see they are .jpeg. You probably have Stable Diffusion set to compress images automatically, which gives you both png and jpeg, but you uploaded the jpeg.
The prompts can still be read if you download the images, but just not on the civitai website.
For some reason, the 4 images you uploaded 2 days ago (which are also jpeg) actually show the prompt.
Maybe you clicked something wrong when uploading the first batch of images?
@upscaleanon537 I find the reaseon.The model directory contains Chinese words.
not for img2img unfortunately
Not directly img2img, but you can use controlnet to lock composition and object proportions of a reference image. Combined with the fitting prompt, that gives even better results (with only a bit more manual work)
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.