Download
1 variant available
460 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
10 1 2 3 4 5 6 7 8 9
20 1 2 3 4 5 6 7 8 9
(10)
Aug 25, 2026

3530 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
3K0 1 2 3 4 5 6 7 8 9K
License:
AnimaThe 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
A LoRa for the character "Iro Kawase" out of the game "crush crush"
offers both her normal outfit and her bikini
Anima:
I recommend a strength of 0.8 - 1 | all images are generated with 1
highly recommend using highres-fix for better quality
same data set as the other versions just had to add some tags for better learning.
Still trying to figure out the best settings for Anima LoRa training so they might wary in quality.
Looking for feedback and ideas.
PonyXL:
I recommend a strength of 0.8 - 1
highly recommend using highres-fix for better quality
mainly tested on prefectponyXL
Trained as i recently started using PonyXL and someone asked for a updated version
I still dont know where my LoRas are quality wise so i am looking for feedback and hope people enjoy using them.
trained locally so it takes me a while to train each of them. God bless my new GPU.
OLD SD1.5
for this lora i recommend a strength of between 0.65 - 0.85 but you can try around with more or less. (most images were generated with 0.75)
Like always i recommend using highres. fix for better quality and details.
For full body pictures of her main outfit simply add the tag "sneakers" for her shoes
if you enjoy my LoRas then please leave a like and let me know if something needs fixing.
I am also happy about every comment and picture posted using my LoRas as i like seeing what other ppl manage to do with them and how they hold up.
