Type | |
Stats | 1,148 |
Reviews | (136) |
Published | Mar 23, 2023 |
Base Model | |
Training | Epochs: 10 |
Trigger Words | combine soldier |
Training Images | Download |
Hash | AutoV2 9111452A93 |
Overwatch, 742 at grid 9-7. Holding for advanced directive.
This is a lora trained on 24 hand picked images of Combine Soldiers from Half Life 2. It was trained mostly on digital art, but there are some 3d renders in there for variety. It was kind of hard finding quality art of just combine soldiers online, so there are only 24 images in this set. I was also planning on having this be trained on both combine soldiers and combine elite soldiers, but changed my last last minute. It was trained on danbooru-style tags, so use those when trying to prompt - I've included the tags in a .txt file in the "Training Images" download below.
It works pretty well with most .ckpts, but I recommend Anything 4.5 or AOM3A1B, I used them for the example images.
NOTES:
-Make sure to specify the color of the glowing eye-holes, I forgot to do that for most of the example images so sometimes different colors show up. It should be able to do red, or blue colored easily natively, but also other colors if you emphasize it enough.
-A good amount of images were trained with characters holding guns, so its slightly better at doing that than usual.
-The armor vest can be a little iffy, and sometimes it won't look exactly like the combine armor usually does.
Example prompt: (See example images, all images should still have their metadata. Just drag the images onto the PNG info tab and import it.)