Sign In

Download

1 variant available

fp32 SafeTensor

monstyle.safetensors

Full precision, largest file • 175.1 MB

Verified:

Type
LoRA
Stats

212

1

Reviews
Published

Aug 9, 2026

Base Model

Anima

Training
Steps: 30
Epochs: 5
Usage Tips
Strength: 0.4
Hash
AutoV2
BE1CC3C8A7
default creator card background decoration
Followers - 9

9

Likes - 29

29

License:

Anima

The Anima Model is licensed by CircleStone Labs LLC. Copyright CircleStone Labs LLC. IN NO EVENT SHALL CIRCLESTONE LABS LLC BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

Built on NVIDIA Cosmos

1.png

Set the LoRA strength to around 0.4!!!

The trigger word is monstyle.

I used the ER SDE sampler for the example images.

This is my first LoRA, so I may have overtrained it a bit xD. Go much above 0.4 and you can expect some good old-fashioned hand and finger horror, just like in the glorious SDXL days.

Any feedback is very welcome. Also, feel free to leave a reaction if you like it. I’d love to start earning some Blue Buzz so I can eventually train LoRAs directly on Civitai instead of torturing my poor old RTX 3060. This one took around 4 hours to train, although I’m pretty sure I messed up the RAM/VRAM swapping settings. With better settings, I suspect I could’ve gotten it down to around 2 hours.

And since you somehow made it this far, I also highly recommend checking out the Kaibashira style:

https://civitai.com/models/841662/artstyle-or-kaibashira

In my totally subjective opinion, it’s probably the best art style you can download on Civitai.