Download
1 variant available
270 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
00 1 2 3 4 5 6 7 8 9
(4)
Aug 9, 2026
Improved the dataset and switched to a new model architecture.
Training Information (Anima DiT)
Output Name: bf16_Anima_V1_Nyanko
Network Module: networks.lora_anima
Epochs: 12
Max Train Steps: Auto (Epoch)
Learning Rate: 0.0001
Text Encoder LR: 5e-05
Optimizer: AdamW8bit
Mixed Precision: bf16
LR Scheduler: constant
LR Warmup Steps: 0
Rank: 8/8 (Dim, Alpha)
### Anima Arguments:
+ timestep_sample_method: logit_normal
+ discrete_flow_shift: 3
+ weighting_scheme: logit_normal
### Dataset Info (dataset.toml):
+ Batch Size: 2
+ Resolution: 1280x1280
+ Steps (calculated): 276
+ Bucketing: True
### Datasets Parameter:
+ 2_pink pikadot : images: 14 | repeats: 2 | keep_tokens: 1 | flip_aug: False | shuffle_caption: False
+ 2_maid : images: 6 | repeats: 2 | keep_tokens: 1 | flip_aug: False | shuffle_caption: True
+ 1_chibi : images: 5 | repeats: 1 | keep_tokens: 1 | flip_aug: False | shuffle_caption: True
Show more

960 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
2760 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
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
Tweak:
that you can tweak a few character details to make them unique if you wish.
Ear style (original):
no fur cat earsOr style 2:
animal ears, fur cat ears, animal ear fluff,Eyes color:
black eyesOr
brown eyesNote:
Aug 17, 2025 Today I decided to make a LoRA for Nyako next, because I really want to finish the Wanko–Nyako pair. Since I was in a bit of a rush, I didn’t check the dataset too carefully, so some errors will show up — the most detailed one being the green clover with the coccinella (ladybug), which sometimes appears and sometimes doesn’t in certain images. It was trained using the same settings as Wanko Pony, with 17 images for each of them, and the training was done on Colab with Pony v6 as the base. As for Illustrious, I trained that one afterward.
`Aug 9, 2026` It’s been a few months, and I’ve decided to create an "Anima" version to align with the WanWan updates and establish the Anima model as the default standard for all future LoRAs; however, if you would like an "Illustrious" version, just leave a comment and I’ll make it.
Prompt Trigger:
Base Character Style 1:
ny4nk0, 1girl, solo, catgirl, animal ears, fur cat ears, animal ear fluff, brown eyes, white hair, short hair, low twintails, hair between eyes, hair bobbles, hair ornament, green clover ornament, blush stickers, cat tail, Base Character Style 2:
ny4nk0, 1girl, solo, catgirl, no fur cat ears, black eyes, white hair, short hair, low twintails, hair between eyes, hair bobbles, hair ornament, green clover ornament, blush stickers, cat tail,Outfit Tank Top:
tank top, pink polka dot on tank top, pink bow on tank top, collarbone, blue skirt, white kneehighs,Outfit Maid:
black maid, puffy short sleeves, white apron, long apron, black bow ornament, white bow neck, white bow behind, white socks, white kneehighs, black maid outfit, bow hair ornament