I had more than 10 attempts and training sessions in making v3.
Use keywords loona and hellhound in your prompt.
Bonus:
Loona has a unique set of clothing. Added two unique tags for her top wear and bottom wear to make them more distinct. Following are tags for her clothing set.
---------------| helper tagsloonacroptop | crop top, shirt, topwearloonashorts | shorts, bottomwear
To get the strings of her crop top nicely, you can add pentagram in the prompt. Vary the emphasis, () or [], until you get them right.
There were images of her without the crop top but with the strings, which means you can have her crop top but without the strings by specifying pentagram in the negative prompt.
Others:toeless legwear, thigh highs, stockings, legwearfingerless gloves (should be fingerless, not just gloves)ear piercing, ear ring eyebrow piercing, eyebrow ring notched ear
These are the tags for her red eyes with white iris and slit pupils in the dataset.slit pupils, white eyes, red sclera
Because of the added custom tags, you can make other characters wear her clothes as well (see my comment)
Also, check out my Twitter profile, I post high-quality furry AI art:
https://twitter.com/pika_fluff_tuft
Follows are much appreciated ^^
Description
v3
Improved pretty much in all aspects
FAQ
Comments (16)
Hi, add the keyword for legs please. If you make it realistic, then the legs are like a girl's.
Could you elaborate more on that? I wrote a description regarding her full clothing set in the description. Click the "show more" button to see the rest of it. If you're having trouble with making the legs look furry, then I suggest experimenting with various phrases like "detailed fluffy fur texture". Check out the posts of the other people here. You can get the prompt they used by clicking the info button at the lower right of the image.
@pika__ I did this on the Deliberate model, and if the moon is standing at full height, it has human legs
Lmao make a freddy fazbear one
Nah fredina
Hi, I'm new to this, how do I use the data like "promp", negative promp", "model"you use in my stable diffusion ?
Hello, I use automatic1111 webui for stable diffusion.
https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features
If this is your first time using stable diffusion I suggest asking in a stable diffusion discord server or the stable diffusion Reddit community.
This is hands down my favorite lora. Works on just about any model I can throw at it genning anything from line art to photo realism. 10/10
I just.... I love the future ya'll. Bless you Pika this is an incredible model thanks for putting the work in and literally moving mankind forward 🤣
Excellent art
What is this trained on? This doesn't seem to be usable with the AiHorde, is it based on sd 1.5,2.0,2.1?
It is based on sd1.5 model type
Ahhhh @pika__ I've just realized the problem, this is marked as Other as base model instead of sd1.5. That makes it incompatible with the horde afaik, because it reads it won't work with the selected model.
Can you make a Lora for Charlie Morningstar please?
Only if you got 50k to drop on it (✿◡‿◡)
Nice
Details
Files
loona-lora-v3-latest.safetensors
Mirrors
loona-lora-v3-latest.safetensors
loona-lora-v3-latest.safetensors
Loona_HB.safetensors
Ln_HB.safetensors
loona-lora-v3-latest.safetensors
loonaHelluvaBossLora_loonaLoraV3Latest.safetensors
loona-lora-v3-latest.safetensors
wlf.safetensors
ZQZPVklBB2Ngkp51q00sM (1).safetensors
ZQZPVklBB2Ngkp51q00sM.safetensors
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.









