Download
1 variant available
bf16 SafeTensor
wagging_v0.0.0_0900_cosine-200.safetensors
BF16, good balance • 66.28 MB
Verified: 15 hours ago
1010 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
00 1 2 3 4 5 6 7 8 9
(15)
Aug 25, 2026
Initial.
dim16/alpha16, AdamW8bit, UNet LR 1e-04 1100steps(constant 900 + cosine 200/300), 512² resolution, 60 images.

6570 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
4.3K0 1 2 3 4 5 6 7 8 9.0 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
This LoRA is designed to generate a single, clearly defined afterimage for a moving body part, capturing the motion as a distinct visual detail. The target body parts include ears, arms, wings, and tails.
For example, when depicting an arm being raised as part of a waving gesture, the LoRA generates a single afterimage of the arm in its lowered position.
For v0.0.0, the LoRA was trained on a total of 60 training images: 10 each for ears (ear wiggle), one hand (waving), both arms (flailing), wings (flapping), tails (tail wagging), and hair (hair wagging).
However, the training for hair was unsuccessful. In particular, it cannot reproduce the classic ahoge bouncing/pulsing motion.
Tragic. 😭 uoh

